Control device of ultrasonic physiotherapy instrument and ultrasonic physiotherapy instrument

Abstract

The invention discloses a control device of an ultrasonic physiotherapy instrument and the ultrasonic physiotherapy instrument, and relates to the technical field of physiotherapy devices. Three kinds of heterogenous data of point cloud data of space geometric information, thermal imaging data of body surface temperature information and medical images of image information are processed, mapping processing is carried out to obtain fusion information, and three-dimensional information reflecting the individual anatomy and winning state of a patient is constructed. And determining a preset physiotherapy path based on the fusion information and the corresponding constraint condition, and determining the physiotherapy path of the current time in real time through the preset physiotherapy target region and the current fusion information, thereby ensuring that stable and effective energy output can be given to the patient in each physiotherapy. The path parameters of the current physical therapy path are adjusted to obtain the adjusted current physical therapy path, the information corresponding to the individualized physiological state of each patient is mapped to obtain the fusion data of the patient, the subjective experience of people is abandoned, the deep focus is focused through the multi-modal data, and the accuracy is improved.

Description

Technical Field

[0001] This application relates to the field of physiotherapy device technology, and in particular to a control device for an ultrasonic physiotherapy device and an ultrasonic physiotherapy device. Background Technology

[0002] Traditional manually operated ultrasound therapy devices consist of a main unit and a movable treatment head. They rely entirely on the therapist's subjective experience, with the position, angle, and distance determined by visual observation and touch. This results in dispersed energy and poor precision in targeting deep lesions. Furthermore, the traditional blind operation mode involves a crude dosage of fixed power and time, rather than parameter settings tailored to the patient's individual physiological condition, reducing therapeutic effectiveness. Moreover, this heavy reliance on patient complaints and therapist judgment poses a risk of overheating and burns to patients with reduced sensitivity or unclear communication, compromising safety.

[0003] Therefore, how to achieve precise physical therapy for each patient while ensuring safety is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] The purpose of this application is to provide a control device for an ultrasonic physiotherapy device and an ultrasonic physiotherapy device, so as to solve the problems of poor physiotherapy effect, poor accuracy for deep lesions and reduced safety of conventional physiotherapy devices.

[0005] To solve the above-mentioned technical problems, this application provides a control device for an ultrasonic physiotherapy instrument, comprising: The first acquisition module is used to acquire spatial geometric information, body surface temperature information and image information of the area to be treated, and perform mapping processing to obtain fused information; The first determining module is used to determine a preset physiotherapy path based on the fusion information and the corresponding constraints. The second determining module is used to identify and process the body surface temperature information within the preset physiotherapy path to obtain a preset physiotherapy target area; and to determine the current physiotherapy path based on the preset physiotherapy target area and the fusion information. The first adjustment module is used to adjust the path parameters of the current physiotherapy path to obtain the adjusted current physiotherapy path, so that the execution component can perform the processing to complete the current control processing.

[0006] On one hand, the first acquisition module includes: The first generation module is used to scan the area to be treated using a three-dimensional scanning method to generate three-dimensional point cloud information to form a physiotherapy reference coordinate system; wherein, the three-dimensional point cloud information includes the sensitive areas corresponding to the bone tissue structure and key risk anatomical structures of the area to be treated. The second generation module is used to generate a temperature data heat map by scanning a preset area corresponding to the area to be treated with an infrared thermal imager, and attach it to the position corresponding to the three-dimensional point cloud information. The first extraction module is used to extract feature information from the attached 3D point cloud information and the image information using an iterative nearest-point algorithm, and to iteratively output a rotation matrix; The first transformation module is used to transform the image information into the physiotherapy reference coordinate system according to the rotation matrix to obtain the fused information.

[0007] On the other hand, the first determining module includes: The first output module is used to pre-call the pre-trained model and input the image information of the fused information to output a set of key point coordinates; The third determining module is used to determine the theoretical coordinate information of the target acupoint based on the bone measurement method and the set of key point coordinates, so as to determine the initial acupoint candidate list. The first correction processing module is used to correct the preset target coordinates based on the body surface temperature information of the initial acupoint candidate list and the fused information to obtain the corrected target coordinates. The fourth determining module is used to determine the preset physiotherapy path based on the target coordinates and the constraints.

[0008] On the other hand, the fourth determining module includes: The first acquisition module is used to acquire the sensitive region corresponding to the fused information; The fifth determining module is used to determine the first straight-line path from the entrance of the ultrasonic physiotherapy head to the target point coordinates; The first module is used to designate the intersecting area as a risk area when the first straight path intersects with the sensitive area. The sixth determining module is used to determine a first preset area, centered on the risk area, as an obstacle avoidance constraint area; The seventh determining module is used to determine the depth information between the entrance and the target point coordinates; The eighth determining module is used to determine the preset physiotherapy path based on the relationship between the obstacle avoidance constraint area and the risk area, and the relationship between the depth information and the depth constraint information.

[0009] On the other hand, the second determining module includes: The second extraction module is used to extract potential physiotherapy target areas corresponding to body surface temperature information that are higher than preset body surface temperature information from the body surface temperature information. The first screening module is used to screen out potential physiotherapy target areas corresponding to unobstructed sound beam paths from the entrance to the potential physiotherapy target area, so as to use them as the first target physiotherapy target areas. The second screening module is used to screen out a second target physiotherapy target area that is at a preset distance from the sensitive area within the first target physiotherapy target area, and to use the second target physiotherapy target area as the preset physiotherapy target area.

[0010] On the other hand, the second determining module includes: The third generation module is used to generate an ultrasonic sound field distribution based on different preset distances between the preset physiotherapy target area and the sensitive area. The ninth determining module is used to determine the target electrode plate spacing based on the location information, depth information, and relative position information of the preset physiotherapy target area and the sensitive area in the ultrasonic sound field distribution. The tenth determining module is used to calculate the surface acoustic intensity in reverse based on the depth information and the tissue average attenuation coefficient, so as to determine the output power to be achieved. The eleventh determining module is used to determine the current physiotherapy path based on the shape of the preset physiotherapy target area, the spacing between the target electrode plates, and the output power.

[0011] On the other hand, the first adjustment module includes: The first receiving module is used to receive adjustment instructions through an interactive interface; The second adjustment module is used to adjust the path parameters of the current physiotherapy path according to the adjustment instruction to obtain the adjusted current physiotherapy path; The adjustment instruction is one of at least one of the following: Instructions for rotating, scaling, and slicing the fused information accordingly; An instruction to adjust the target area of ​​the current physiotherapy pathway; An instruction to adjust the energy intensity of the target area of ​​the current physiotherapy pathway.

[0012] On the other hand, following the first adjustment module, it also includes: The return module is used to take the adjusted current physiotherapy path as the new preset physiotherapy path and return to the preset physiotherapy target area obtained by identifying and processing the body surface temperature information, so as to perform the next physiotherapy treatment until the physiotherapy duration reaches the preset duration and the physiotherapy control ends.

[0013] On the other hand, it also includes: The second acquisition module is used to acquire the fusion information after physiotherapy; The first analysis module is used to perform quantitative analysis based on the fusion information before and after physiotherapy to determine the information change indicators. The fourth generation module is used to generate an evaluation report based on the information change indicators.

[0014] To address the aforementioned technical problems, this application also provides an ultrasonic physiotherapy device, comprising a controller, an execution component, and a treatment head; wherein the execution component is a robotic arm, and the treatment head includes a sensing module; the sensing module is connected to the controller; the controller is connected to the robotic arm; and the robotic arm is connected to the treatment head. The controller includes the control device of the ultrasonic physiotherapy device described above.

[0015] This application provides a control device for an ultrasonic physiotherapy instrument. First, it maps and processes the collected spatial geometric information, surface temperature information, and image information of the area to be treated to obtain fused information. It utilizes diversified acquisition of multimodal data, simultaneously processing point cloud data (spatial geometric information), thermal imaging data (surface temperature information), and medical image data (image information). The goal of this fusion is not to generate a surface path, but to construct three-dimensional information reflecting the patient's individual anatomy and physical condition, providing a foundation for subsequent energy field simulation and in vivo target identification. This represents a qualitative leap from surface navigation to in vivo mapping and prediction. It also achieves a leap from surface projection to precise three-dimensional positioning of the physiotherapy target area, improving positioning accuracy to the millimeter level. Second, based on the fused information and corresponding constraints, a preset physiotherapy path is determined. Path planning considers not only the geometry of the heterogeneous data but also the ultrasonic sound field simulation, thereby generating an optimized path that maximizes energy absorption in the target area and minimizes exposure to non-target areas. Secondly, within the preset physiotherapy path, the surface temperature information is identified and processed to obtain the preset physiotherapy target area. Based on the initial optimized path described above, during the actual physiotherapy process, it is necessary to analyze the thermal imaging data of the surface temperature information in real time to automatically identify potential physiotherapy target areas. The thermal anomaly areas are compared and verified with the anatomical structure of the fusion information to determine the preset physiotherapy target area that ultimately needs physiotherapy. Then, based on the preset physiotherapy target area and the fusion information, the current physiotherapy path is determined. Compared with the conventional approach of using a fixed power and time-based coarse dosage mode, this application further determines the current physiotherapy path through the fusion information, which can establish an objective, real-time, and continuous temperature monitoring and active safety protection mechanism, overcoming user subjective judgment and fundamentally eliminating the risk of overheating and burns, especially ensuring the physiotherapy safety of patients with impaired sensation. By determining the current physiotherapy path in real time through the preset physiotherapy target area and the current fusion information, the individualization, precision, and repeatability of the physiotherapy dosage can be achieved, ensuring that each physiotherapy provides the patient with a stable and effective energy output. At the same time, the ultrasound energy is precisely focused on the deep lesion target point, avoiding excessive stimulation of non-target tissues. Finally, the path parameters of the current physiotherapy path are adjusted to obtain the adjusted current physiotherapy path. Considering that the current physiotherapy path is automatically generated during the data processing stage of the physiotherapy device, the rationality of the path is verified from different perspectives through interactive operation. While adjusting the path parameters of the current physiotherapy path, the final current physiotherapy path is generated, which can then be executed by the execution component to complete the current control processing. The entire process maps information corresponding to each patient's individual physiological state to obtain their own fused data, abandoning subjective human experience and focusing on deep lesions through multimodal data to improve accuracy. The physiotherapy path obtained based on multimodal data processing improves both safety and therapeutic effect.

[0016] In addition, this application also provides an ultrasonic physiotherapy device that has the same beneficial effects as the control device of the ultrasonic physiotherapy device described above. Attached Figure Description

[0017] To more clearly illustrate the embodiments of this application, the accompanying drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A structural diagram of a control device for an ultrasonic physiotherapy instrument provided in an embodiment of this application; Figure 2 A structural diagram of an ultrasonic physiotherapy device provided in an embodiment of this application; Figure 3 A flowchart illustrating a control method for an ultrasonic physiotherapy device provided in this application embodiment; Figure 4 A structural diagram of a control device for another ultrasonic physiotherapy device provided in an embodiment of this application; Figure 5 A flowchart illustrating another control method for an ultrasonic physiotherapy device provided in this application embodiment. Detailed Implementation

[0019] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0020] The core of this application is to provide a control device for an ultrasonic physiotherapy device and an ultrasonic physiotherapy device, in order to solve the problems of poor physiotherapy effect, poor accuracy for deep lesions, and reduced safety of conventional physiotherapy devices.

[0021] To enable those skilled in the art to better understand the present application, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0022] Currently, the technology available on the market is the traditional manually operated ultrasonic physiotherapy device, which mainly consists of a main unit and a movable treatment head (electrode plate) connected by a cable. During treatment, it relies entirely on the experience and feel of the therapist: the therapist manually places the treatment head near or in contact with the patient's skin, adjusts the relative position between the two treatment heads and the distance from the skin through visual inspection and experience, and sets fixed treatment power and time parameters.

[0023] Conventional physiotherapy devices are open-loop control systems that rely on subjective human experience and lack objective feedback. The physiotherapy process lacks precise spatial navigation. The position, angle, and distance of the treatment head depend entirely on the therapist's visual observation and touch, making it impossible to ensure precise energy focus on deep lesions. This leads to dispersed energy, significantly reduced efficacy, and potential unnecessary stimulation of non-target tissues (such as bone). The setting and execution of the physiotherapy dose (energy) are based on fixed empirical parameters, rather than the patient's individual physiological state. The system cannot sense actual temperature changes in tissues during treatment, thus failing to achieve real-time dose adjustment based on biofeedback. This results in poor consistency of treatment doses, making it difficult to guarantee and quantify efficacy. Safety is highly dependent on the patient's complaints and the therapist's subjective judgment. For patients with impaired sensation or unclear expression, there is a significant risk of overheating and burns. The lack of objective, continuous temperature monitoring and proactive safety intervention mechanisms in current technology is a significant safety hazard in clinical practice. The entire physiotherapy process lacks objective, quantitative data recording. The inability of physical therapists to provide reports on key parameters such as energy deposition and temperature changes hinders reliable data support for efficacy evaluation, medical record management, and clinical research, thus impeding the optimization and standardization of physical therapy protocols. The control device for the ultrasonic physical therapy instrument provided in this application can solve the aforementioned technical problems.

[0024] Figure 1 A structural diagram of a control device for an ultrasonic physiotherapy instrument provided in an embodiment of this application is shown below. Figure 1 As shown, it includes: The first acquisition module 11 is used to acquire spatial geometric information, body surface temperature information and image information of the area to be treated, and perform mapping processing to obtain fused information; The first determining module 12 is used to determine the preset physiotherapy path based on the fused information and the corresponding constraints. The second determining module 13 is used to identify and process body surface temperature information within the preset physiotherapy path to obtain a preset physiotherapy target area; and to determine the current physiotherapy path based on the preset physiotherapy target area and fusion information. The first adjustment module 14 is used to adjust the path parameters of the current physiotherapy path to obtain the adjusted current physiotherapy path, so that the execution component can complete the current control process.

[0025] Specifically, acquiring spatial geometric information of the treatment area can be achieved by scanning specific body areas of the patient receiving treatment using a 3D depth camera. The acquired data on the shape, contour, surface structure, and spatial location of the treatment area can include point cloud data, depth maps, surface mesh models, spatial pose and position information, and, in the case of dynamic capture, deformation and dynamic geometric information. Point cloud data, containing 3D coordinates, constitutes discrete sampling of the surface of the treatment area, reflecting the true shape of the treatment area (such as tumor surface projection, wounds, radiotherapy target areas, etc.). Depth maps represent the distance from each pixel to the camera in image form and can be used to reconstruct surface height changes, such as chest wall undulations and limb curvature. Surface mesh models, generated from point clouds, are triangular patches or curved meshes that form a continuous 3D surface, which can be used for visualization, volume calculation, or registration with a planning system. Spatial pose and position information refers to the absolute or relative position of the treatment area in the camera coordinate system (or global coordinate system). For radiation therapy or surgical navigation, precise spatial relationships between the lesion and equipment (such as linear accelerators) are required. Deformation and dynamic geometry information, including dynamic deformation data of the surface of the treatment area during breathing, heartbeat, or changes in body position, is crucial for precise radiotherapy (such as gated radiotherapy).

[0026] Unlike traditional two-dimensional data, three-dimensional depth cameras directly provide spatial distance information, enabling the reconstruction of 3D structures with true scale, which is crucial for medical applications that require precise positioning, measurement, or registration.

[0027] Body surface temperature information is temperature data synchronously acquired by an infrared thermal imager. Image data is data acquired during medical image acquisition by computed tomography (CT) and magnetic resonance imaging (MRI).

[0028] After data acquisition, preprocessing is required before mapping. For example, spatial geometric information is processed using point cloud algorithms to construct a 3D surface model. Body surface temperature information is analyzed using temperature analysis algorithms to generate a thermal distribution map. Noise reduction and filtering are applied to the spatial geometric information, and non-uniformity correction is performed on the body surface temperature information to prepare for subsequent fusion. It should be noted that conventional solutions primarily focus on visual positioning and trajectory tracking, with image acquisition followed by image processing to generate body surface motion trajectories. However, this application aims for a deeper level of multivariate data processing. Instead of generating surface paths, it constructs a model reflecting the individual patient's anatomy and physiological state, providing a foundation for subsequent energy field simulation and in vivo targeting—a qualitative leap from surface navigation to in vivo mapping and prediction.

[0029] The process of mapping three types of data to obtain fused information essentially transforms two-dimensional data into three-dimensional data, enabling the leap from surface projection to precise three-dimensional positioning of the therapeutic target area, thus achieving millimeter-level accuracy. The coordinate systems corresponding to the three types of data are unified into a single coordinate system to obtain fused information. This coordinate system unification can be achieved through pairwise registration and step-by-step mapping, without limitation, and can be set according to actual conditions. Furthermore, coordinate system unification can be achieved by constructing a digital twin model. For example, using registered CT / MRI data as a reference, image segmentation algorithms (such as the U-shaped Convolutional Neural Network for Segmentation, U-Net) can be used to automatically segment different tissues such as bone, muscle, fat, and the target area, generating a separate three-dimensional mesh model for each type of tissue. For each tissue region in the model, the system retrieves and assigns acoustic properties from a biophysical database, such as sound velocity, density, acoustic impedance, and ultrasonic attenuation coefficient. Simultaneously, surface temperature properties are also associated with the corresponding skin mesh. All tissue models with attributes, as well as skin models with temperature attributes, are integrated within a unified coordinate system and encapsulated into a multi-scale, multi-physics digital twin. This twin is not only visual but also computationally and simulation-capable. It can realistically simulate how ultrasound waves propagate, reflect, attenuate, and generate heat within a specific patient's body.

[0030] Based on fused information and corresponding constraints, a preset physiotherapy path is determined. This path planning considers not only geometry but also integrates ultrasonic sound field simulation based on the finite element method to predict energy deposition distribution at different positions and postures, thereby generating an optimized path that maximizes energy absorption in the target area and minimizes exposure to non-target areas. Path planning is not based on a single dataset; the physiotherapy path must avoid sensitive structures such as bones and nerves marked in the aforementioned data. This ensures that ultrasonic energy covers and acts on the target area (e.g., inflamed areas) to the maximum extent and uniformly. The planning algorithm pre-sets multiple candidate scanning paths for the ultrasonic physiotherapy head in the model. Then, for each path, the system performs rapid ultrasonic sound field simulation to predict the energy distribution within the tissue. The algorithm evaluates which path optimally achieves the physiotherapy goal (e.g., highest and most uniform sound intensity in the target area) while meeting safety constraints. The algorithm outputs the optimal set of parameters, including the robotic arm motion path, physiotherapy head posture, sound intensity, and work cycle. During physiotherapy, the robotic arm will strictly move along this path to ensure that energy acts on the body in the predicted manner. The path generation here can be based on the digital twin model in the above embodiments to output the path, or it can be based on a smooth curve or acupoint information. For the smooth curve approach, a broken line is formed by directly connecting the marked key therapy points, or the speed of the device moving along the curve is smoothed. Specifically, interpolation calculation is first performed, and natural cubic splines are used to fit discrete points to form parametric equations. Then, the curvature distribution of the entire curve is calculated to check if there are curves with excessive curvature that the device cannot follow. Then, constraint parameters are set, such as maximum moving speed, maximum acceleration, maximum jerk, and target energy density, to obtain the discrete time. Then, a preset therapy path is generated based on these constraints.

[0031] The process of determining the preset physiotherapy path for acupoint information involves suggesting a relative coordinate system based on human posture estimation and key point extraction to eliminate positioning errors caused by changes in patient position and body shape. Then, based on anatomical mapping and a preliminary list of candidate acupoints, thermal imaging physiological correction is applied. Before generating the path, anatomical constraints are verified for each target area, corresponding to constraints such as obstacle avoidance, depth, and incident angle constraints.

[0032] Within a preset physiotherapy path, surface temperature information is processed to obtain a preset physiotherapy target area. Upon initially obtaining the preset physiotherapy path, the next preset physiotherapy target area is identified in real-time based on the target points within that path. If the real-time acquired parameters deviate, the corresponding next preset physiotherapy target area will also deviate from the next target area corresponding to the preset physiotherapy path. Therefore, the corresponding preset physiotherapy target area is identified based on the first few target points of the preset physiotherapy path and the real-time acquired surface temperature information. This facilitates subsequent determination of the current physiotherapy path with fusion information, achieving a scanning strategy of gradually narrowing the physiotherapy path. The specific algorithm used for gradually narrowing the physiotherapy path can be spiral contraction, concentric contraction, or focused scanning; there are no limitations, and it can be planned according to the actual situation. However, each determination of the physiotherapy path is based on real-time fusion information. The preset physiotherapy target area is identified based on the current surface temperature information; the current preset physiotherapy target area will change from the previous preset physiotherapy target area, gradually narrowing.

[0033] The current physiotherapy path in this application is determined and identified for each current physiotherapy path in the closed-loop process to ensure that the physiotherapy path changes with real-time fusion information.

[0034] The adjustment of the path parameters of the current physiotherapy path is made in consideration of the automated calculation method used in the above embodiment. The path parameters are then modified through interactive operations where the user can control whether adjustments are needed, verifying the path's rationality from different perspectives. This modification is then executed by the actuator. The actuator here is a robotic arm, which automatically positions itself to control the physiotherapy initiation of the treatment head.

[0035] This application provides a control device for an ultrasonic physiotherapy instrument. First, it maps and processes the collected spatial geometric information, surface temperature information, and image information of the area to be treated to obtain fused information. It utilizes diversified acquisition of multimodal data, simultaneously processing point cloud data (spatial geometric information), thermal imaging data (surface temperature information), and medical image data (image information). The goal of this fusion is not to generate a surface path, but to construct three-dimensional information reflecting the patient's individual anatomy and physical condition, providing a foundation for subsequent energy field simulation and in vivo target identification. This represents a qualitative leap from surface navigation to in vivo mapping and prediction. It also achieves a leap from surface projection to precise three-dimensional positioning of the physiotherapy target area, improving positioning accuracy to the millimeter level. Second, based on the fused information and corresponding constraints, a preset physiotherapy path is determined. Path planning considers not only the geometry of the heterogeneous data but also the ultrasonic sound field simulation, thereby generating an optimized path that maximizes energy absorption in the target area and minimizes exposure to non-target areas. Secondly, within the preset physiotherapy path, the surface temperature information is identified and processed to obtain the preset physiotherapy target area. Based on the initial optimized path described above, during the actual physiotherapy process, it is necessary to analyze the thermal imaging data of the surface temperature information in real time to automatically identify potential physiotherapy target areas. The thermal anomaly areas are compared and verified with the anatomical structure of the fusion information to determine the preset physiotherapy target area that ultimately needs physiotherapy. Then, based on the preset physiotherapy target area and the fusion information, the current physiotherapy path is determined. Compared with the conventional approach of using a fixed power and time-based coarse dosage mode, this application further determines the current physiotherapy path through the fusion information, which can establish an objective, real-time, and continuous temperature monitoring and active safety protection mechanism, overcoming user subjective judgment and fundamentally eliminating the risk of overheating and burns, especially ensuring the physiotherapy safety of patients with impaired sensation. By determining the current physiotherapy path in real time through the preset physiotherapy target area and the current fusion information, the individualization, precision, and repeatability of the physiotherapy dosage can be achieved, ensuring that each physiotherapy provides the patient with a stable and effective energy output. At the same time, the ultrasound energy is precisely focused on the deep lesion target point, avoiding excessive stimulation of non-target tissues. Finally, the path parameters of the current physiotherapy path are adjusted to obtain the adjusted current physiotherapy path. Considering that the current physiotherapy path is automatically generated during the data processing stage of the physiotherapy device, the rationality of the path is verified from different perspectives through interactive operation. While adjusting the path parameters of the current physiotherapy path, the final current physiotherapy path is generated, which can then be executed by the execution component to complete the current control processing. The entire process maps information corresponding to each patient's individual physiological state to obtain their own fused data, abandoning subjective human experience and focusing on deep lesions through multimodal data to improve accuracy. The physiotherapy path obtained based on multimodal data processing improves both safety and therapeutic effect.

[0036] In some embodiments, the first acquisition module includes: The first generation module is used to scan the area to be treated using a three-dimensional scanning method to generate three-dimensional point cloud information to form a physiotherapy reference coordinate system; wherein, the three-dimensional point cloud information includes the sensitive areas corresponding to the bone tissue structure and key risk anatomical structures of the area to be treated. The second generation module is used to generate a temperature data heat map by scanning the preset area corresponding to the area to be treated with an infrared thermal imager, and attach it to the position corresponding to the three-dimensional point cloud information. The first extraction module is used to extract feature information from the attached 3D point cloud information and image information using the iterative nearest point algorithm, and to iteratively output the rotation matrix; The first transformation module is used to transform the image information into the physiotherapy reference coordinate system according to the rotation matrix to obtain fused information.

[0037] Specifically, during the physiotherapy preparation phase, the patient assumes a comfortable position, fully exposing the area to be treated. The operator turns on the system power, the robotic arm performs a self-check and returns to the zero position, all sensors are activated and calibrated, and the system records the patient's pre-stored medical imaging data.

[0038] Three-dimensional scanning methods, such as using a depth camera (to scan the treatment area along a pre-defined trajectory), are employed to generate three-dimensional point cloud information of the area to be treated. The resolution can be set according to actual conditions, such as 0.5mm. A world coordinate system, i.e., a treatment reference coordinate system, is established. The three-dimensional point cloud information includes key anatomical landmarks, i.e., sensitive areas corresponding to the bone tissue structures and key risk anatomical structures in the area to be treated.

[0039] The infrared thermal imager scans a preset area corresponding to the area to be treated, such as a 5cm radius around the perimeter, with a temperature sensitivity of 0.05℃. It automatically marks areas with abnormal temperatures to generate a temperature data heat map. It shares the field of view with the 3D camera and can be directly and automatically attached to the position corresponding to the 3D point cloud information through built-in calibration parameters to form a skin surface model with temperature attributes.

[0040] An iterative nearest-point algorithm is used to extract feature information from the attached 3D point cloud and image information, iteratively outputting a rotation matrix and translation vector. This algorithm then performs automatic registration, extracting a large number of feature points from both the 3D point cloud and the reconstructed image information (which can be reconstructed from a bone surface model using CT / MRI). Through continuous iteration, an optimal rotation matrix and translation vector are calculated to minimize the average distance between the two point sets. This rotation matrix is ​​then used to rigidly transform the image information into a physiotherapy reference coordinate system established by the 3D camera, obtaining fused information. At this point, the patient's real-time surface morphology, real-time surface physiological state, and high-resolution internal anatomical structures are all precisely aligned in the same 3D space. This allows a hot spot on the skin to accurately correspond to an inflamed area of ​​a specific muscle or joint capsule at a specific depth under the skin.

[0041] The multimodal data fusion provided in this embodiment accurately maps the body surface thermal point cloud to the CT / MRI coordinate system using an iterative closest point algorithm, allowing visualization of the internal structures through the skin. This improves registration accuracy and establishes a rigid connection between real-time body positioning and the treatment path, enabling real-time calculation of the 3D coordinates of the treatment probe relative to internal lesions and achieving real-time navigation. When the patient moves slightly, the position of the treatment head can be dynamically adjusted based on real-time point cloud updates combined with Iterative Closest Point (ICP) transformation. Its rotation matrix serves as the direct input for the execution component's operational control, transforming visual / sensor data into instructions in the execution component's base coordinate system. Based on the fused data, a spiral or curved path that avoids bone and focuses on the lesion is automatically generated, and energy output is automatically adjusted according to the depth of the internal structure, achieving truly intelligent closed-loop physiotherapy.

[0042] In some embodiments, the first determining module includes: The first output module is used to pre-call the pre-trained model and input the image information of the fused information to output a set of key point coordinates; The third determination module is used to determine the theoretical coordinate information of the target acupoint based on the bone measurement method and the set of key point coordinates, so as to determine the initial acupoint candidate list. The first correction processing module is used to correct the preset target coordinates based on the body surface temperature information of the initial acupoint candidate list and the fusion information to obtain the corrected target coordinates. The fourth determination module is used to determine the preset physiotherapy path based on the target coordinates and constraints.

[0043] Specifically, a pre-trained model, such as the YOLOv7-HumanPose model based on the YOLOv7 architecture, is pre-called. Inputting image data, it detects key skeletal points (such as the acromion and anterior superior iliac spine) in real time and outputs a normalized set of keypoint coordinates. This eliminates positioning errors caused by changes in patient position and body shape, establishing a relative coordinate system. Based on the bone measurement method, the theoretical coordinates of the target acupoints are determined using the keypoint coordinate set to establish an initial list of candidate acupoints. Specifically, the bone measurement method calls the `match_acupoint_db(keypoints)` function to output a preliminary list of candidate acupoints. By comparing the thermal distribution around the theoretical coordinates, if significant high or low temperature anomalies are detected, the final target point coordinates are shifted a certain distance towards the center of the thermal anomaly. The coordinates of preset target points are corrected based on the theoretical coordinates of the target acupoints in the initial acupoint candidate list and the body surface temperature information of the fusion information. The principle is to simulate the experience of doctors to find acupoints along the meridians and locate acupoints by heat, so as to ensure that the locked acupoints are living acupoints with physiotherapy value, rather than rigid anatomical points, so as to output an enhanced target point list, that is, the corrected target point coordinates.

[0044] The preset physiotherapy path is determined based on the target coordinates and constraints, including obstacle avoidance constraints, depth constraints, and incident angle constraints. The path optimization algorithm is then run to generate the preset physiotherapy path.

[0045] This embodiment provides a dynamic acupoint recognition algorithm that employs a progressive logic of physiological guidance and anatomical constraint verification to simulate the decision-making process of a physician examining thermography, investigating anatomy, and determining a treatment plan. Its output, an enhanced target point list, serves as input to the path optimization algorithm. The coordinates, depth, and safety constraints in the list represent the therapeutic goals and obstacle avoidance conditions that the optimization algorithm must meet. The core objective of the system is to achieve precise, safe, and personalized physiotherapy. It ensures that energy is only directed to precise locations that are both therapeutically necessary (thermally active), therapeutically possible (anatomically accessible), and safe for treatment (risk avoidance).

[0046] In some embodiments, the fourth determining module includes: The first acquisition module is used to acquire the sensitive areas corresponding to the fused information; The fifth determining module is used to determine the first straight-line path from the entrance of the ultrasonic physiotherapy head to the target point coordinates; The first module is used to identify the intersecting area as a risk area when the first straight path intersects with a sensitive area. The sixth determination module is used to determine a first preset area centered on the risk area as the obstacle avoidance constraint area. The seventh determination module is used to determine the depth information of the entrance and target coordinates; The eighth determination module is used to determine the preset physiotherapy path based on the relationship between the obstacle avoidance constraint area and the risk area, and the relationship between depth information and depth constraint information.

[0047] Specifically, the sensitive areas corresponding to the fusion information are obtained. These sensitive areas include areas surrounding sensitive structures such as nerves, large blood vessels, and organs. The first straight-line path from the entrance of the ultrasonic therapy head to the target coordinates is determined. If this first straight-line path intersects with a sensitive area, the intersecting area is designated as a risk area. Then, a first preset distance from the risk area is determined as the obstacle avoidance constraint area. That is, the first preset area formed by radii of the risk area at the intersection of the paths is the obstacle avoidance constraint area. A preset therapy path is determined based on the relationship between the depth information and depth constraint information between the entrance and the target coordinates. Specifically, the therapy sequence is determined based on acupoint proximity and thermal activity priority; the optimal distance between the execution component or therapy head and each target point is calculated to satisfy the depth constraint information.

[0048] This embodiment provides a preset physiotherapy path determined based on target coordinates and constraints. Traditional straight-line path planning often assumes a uniform or direct penetration of the medium, easily overlooking sensitive tissues (such as superficial nerves, blood vessels, or bony prominences) along the path. This embodiment uses a three-level mechanism—intersection detection, risk area localization, and obstacle avoidance constraint area generation—to forcibly deform a straight path that would otherwise pass through sensitive tissues into a detour path. It not only avoids the sensitive area itself (risk area) but also constructs an additional first preset area (safety buffer zone). This means that even with minor control errors in the robotic arm or slight patient displacement, the sound beam will not touch sensitive tissues, greatly reducing the risk of accidental nerve injury or periosteal overheating. During planning, the system weighs the safety gains of detours against the energy loss caused by path extension. It ensures that the generated preset physiotherapy path meets safety avoidance requirements while remaining within the effective penetration depth of the sound waves, preventing excessive avoidance from causing the energy received by the target area to fall below the physiotherapy threshold.

[0049] In some embodiments, the second determining module includes: The second extraction module is used to extract potential physiotherapy target areas corresponding to body surface temperature information that are higher than the preset body surface temperature information. The first screening module is used to screen out potential physiotherapy target areas corresponding to unobstructed sound beam paths from the entrance to the potential physiotherapy target area, so as to use them as the first target physiotherapy target areas. The second screening module is used to screen out a second target physiotherapy target area that is within a preset distance from the sensitive area within the first target physiotherapy target area, and to use the second target physiotherapy target area as the preset physiotherapy target area.

[0050] Specifically, the system analyzes surface temperature information and automatically identifies areas where the surface temperature is significantly higher than that of surrounding tissues (corresponding to preset surface temperature information). These areas are designated as potential therapeutic target zones. Then, the system checks whether the ultrasound beam path from the entry point to the potential target zone effectively covers the area. In other words, it checks whether the ultrasound beam path from the surface to the potential target zone is unobstructed and whether there are strong reflective or absorptive structures like bones that completely block the path. If not, this area is designated as the first target therapeutic zone. Further screening within the first target zone reveals second target zones that are at a preset distance from sensitive areas. This checks whether the first target zone is too close to critical sensitive structures like bones or nerves to ensure treatment safety. Only areas that simultaneously meet the conditions of anatomical accessibility and distance from critical structures are identified as the final optimized target zones requiring treatment.

[0051] The target region identification algorithm runs as follows: / / Target region identification and optimization algorithm

[0052] FUNCTION identifyAndOptimizeTargets(thermalData, anatomyModel)

[0053] SET potentialTargets=FIND_THERMAL_ANOMALIES(thermalData) / / Find hot anomaly zones

[0054] SET optimizedTargets=[]

[0055] FOR EACH target IN potentialTargets

[0056] IF IS_ANATOMICALLY_ACCESSIBLE(target,anatomyModel)AND

[0057] NOT IS_NEAR_CRITICAL_STRUCTURE(target, anatomyModel) / / Avoid critical structures

[0058] ADD target TO optimizedTargets

[0059] END IF

[0060] END FOR

[0061] RETURN optimizedTargets

[0062] END FUNCTION.

[0063] The preset physiotherapy target area identification process provided in this embodiment is a multi-data source verification process, such as image segmentation algorithms, region growing, threshold segmentation analysis of infrared thermal images, automatically identifying continuous areas with a body surface temperature significantly higher than the surrounding normal tissue, and marking them as suspected inflammation / physiotherapy target areas. The identification of preset physiotherapy target areas is used to eliminate false positive target areas. Accessibility verification is employed: determining whether the ultrasound beam path from the body surface to the suspected target area is unobstructed, and whether there are strong reflective / absorbent bodies such as bones that completely block it. Safety verification is also used: determining whether the suspected target area is too close (e.g., distance <5mm) to important nerves, blood vessels, or other sensitive structures in three-dimensional space. Only areas that simultaneously pass thermal anomaly identification and anatomical accessibility and safety distance verification are determined as optimal physiotherapy target areas, thereby improving the accuracy of preset physiotherapy target area identification.

[0064] In some embodiments, the second determining module includes: The third generation module is used to generate an ultrasonic sound field distribution based on different preset distances between the preset physiotherapy target area and the sensitive area. The ninth determining module is used to determine the target electrode plate spacing based on the location information, depth information, and relative position information of the preset physiotherapy target area and the sensitive area in the ultrasonic sound field distribution. The tenth determination module is used to calculate the surface acoustic intensity in reverse based on the depth information and the average attenuation coefficient of the tissue, so as to determine the output power to be achieved. The eleventh determination module is used to determine the current physiotherapy path based on the shape of the preset physiotherapy target area, the spacing between the target electrode plates, and the output power.

[0065] Specifically, the target region identification process outputs a 3D target region model containing the following key attributes, which are directly used as input conditions for parameter calculation: 1. The three-dimensional geometry and spatial location of the target area: This is the basis for path planning.

[0066] 2. The depth of the target area from the body surface (Depth): This is the core basis for setting the initial power.

[0067] 3. The spatial relationship between the target area and surrounding sensitive tissues (such as bones and nerves): This is the direct reason for determining the safe distance and path avoidance.

[0068] The ultrasonic sound field distribution is generated based on different preset distances between the preset therapeutic target area and the sensitive area. Here, the ultrasonic sound field distribution under different electrode plate spacings is simulated in a virtual environment. In the ultrasonic sound field distribution, the target area position, depth, and relative position of the surrounding bones / nerves are input. That is, based on the preset position and depth information of the therapeutic target area and its relative position with the sensitive area, the sound field cloud map under different spacings is iteratively calculated, and the optimal spacing that maximizes the energy of the target area and keeps the energy of the sensitive area below the safety threshold is output. This determines the target electrode plate spacing, avoids energy defocusing or overheating of bone surface standing waves caused by fixed spacing, and achieves dynamic focusing.

[0069] Furthermore, conventional treatments typically use a fixed power level. However, since each patient's physiological parameters differ, it's necessary to determine the appropriate output power in real-time for each patient to achieve personalized therapy. The surface acoustic intensity is calculated by back-calculating based on depth information and the average tissue attenuation times, thus determining the corresponding output power. The formula is as follows: ; in, The required effective therapeutic sound intensity for the target area. For organizational depth, The average tissue decay coefficient (dynamically adjusted based on the fat / muscle ratio in the digital twin).

[0070] Inversion formula: ; in, For effective radiation area, This refers to the output power.

[0071] The current physiotherapy path is determined based on the shape of the preset physiotherapy target area, the spacing between the target electrode plates, and the output power. The electrode plate spacing determines the focal spot size; the effective focal spot size is fixed for the sound field calculated with the optimal spacing. The scanning step size of the actuator must be less than or equal to a certain percentage (e.g., 50%) of the focal spot diameter to ensure that the scan is complete and the energy is uniform.

[0072] If a large spacing results in focal spot divergence, the path step size needs to be increased; however, since edge energy is low, speed compensation needs to be reduced. If a small spacing results in a sharp focal spot, the path step size needs to be finer to prevent striped treatment blind spots. If the optimal spacing requires the electrode plates to be parallel to the bone surface to avoid reflection, then the path planning of the actuator must not only consider the position but also strictly constrain the orientation angle to ensure that the electrode plates maintain this optimal angle throughout the entire movement.

[0073] Furthermore, since output power and speed are inversely proportional, in high-power scenarios, the actuator must increase its movement speed to prevent excessive energy accumulation per unit area, which could lead to burns. In low-power scenarios, the actuator must decelerate or pause at key points to ensure sufficient dose is delivered to deeper layers. When the path scans to a region of thick fat layer (with high attenuation) marked in the digital twin, the system can simultaneously perform two actions: increasing local power and decreasing local speed. When the path approaches a sensitive area, the system reduces local power and increases throughput (or even skips that small region).

[0074] If the preset therapeutic target area has an irregular shape, with part adjacent to bone and part at greater depth, an optimal spacing needs to be calculated to ensure the sound field covers most of the area, while still posing a small risk of energy leakage near the bone. A baseline power is calculated based on the average depth. In deeper regions, the path planner instructs the robotic arm to decelerate while simultaneously feeding back to the control system to maintain or slightly increase power to compensate for depth attenuation. In near-bone regions, the path planner instructs the robotic arm to quickly sweep across, while simultaneously feeding back to the control system to instantaneously reduce power (or cut off the pulse), using the reduction in the time dimension to compensate for the risk of spatial dimension issues (where spacing cannot be perfectly avoided). Furthermore, if the target area projection is a regular rectangle, a reciprocating scanning path is generated. If the target area projection is circular or irregular, a spiral scanning path is generated to optimize coverage.

[0075] The current physiotherapy path provided in this embodiment may deviate from the preset physiotherapy path. It is a real-time physiotherapy path that determines the optimal electrode spacing to avoid the risk of bone reflexes. Based on depth and tissue attenuation characteristics, it accurately calculates individualized initial power to replace empirical estimates. Combining the target area shape and safety constraints, it plans an adaptive scanning trajectory to achieve intelligent obstacle avoidance and dose uniformity. It transforms from blind operation to precise targeting based on anatomical structure visualization. Through simulation and electronic fences, it actively avoids the risks of neurovascular injury and periosteal overheating.

[0076] In some embodiments, the first adjustment module includes: The first receiving module is used to receive adjustment instructions through an interactive interface; The second adjustment module is used to adjust the path parameters of the current physiotherapy path according to the adjustment instructions to obtain the adjusted current physiotherapy path; The adjustment instruction is one of the following: Instructions for rotating, scaling, and slicing the fused information; Instructions to adjust the target area of ​​the current physiotherapy pathway; An instruction to adjust the energy intensity of the target area of ​​the current physiotherapy pathway.

[0077] Specifically, the physiotherapy paths described in the above embodiments are all automatically generated. Considering that each individual's bone structure, muscle thickness, fat distribution, scar tissue location, and even nerve pathways vary significantly during physiotherapy (especially in pathological conditions, such as organ displacement caused by tumor compression), a virtual reality (VR) device with an interactive interface can visually demonstrate whether the algorithm-generated path traverses areas of anatomical variation unique to the patient (such as abnormal blood vessel pathways). Physicians can manually avoid contraindicated areas not identified by the algorithm (such as recent surgical wounds, implants, and skin abrasions), details that are often only detectable through clinical examination.

[0078] Most automated planning is based on static images (CT / MRI) or static point clouds. However, the human body is soft, and when a patient lies on a physiotherapy bed, the surface and internal organs deform in real time due to gravity, breathing, and muscle tension. While ICP registration can correct rigid body displacement, it struggles to perfectly predict complex soft tissue compression deformation. In VR, physicians can simulate or observe tissue deformation in the patient's actual position. Physicians can dynamically adjust the contact angle and pressure distribution of the physiotherapy head to ensure energy is focused on the actual deformed lesion, not the originally calculated coordinate points. Algorithms can only process objective data (temperature, location, images) and cannot perceive the patient's pain threshold, comfort level, psychological tension, or sudden referred pain. The same energy parameters may be therapeutic for patient A but excruciating pain for patient B. Before or between treatments, physicians can adjust the path, power curve, or dwell time in VR in real time based on the patient's verbal feedback (a slight sting here, insufficient sensation there). Physicians can set more conservative safety boundaries for sensitive areas based on experience, something that cold, impersonal algorithms struggle to quantify.

[0079] Therefore, adjustment commands are received through the interactive interface, and the path parameters of the current physiotherapy path are adjusted based on these commands to obtain the final current physiotherapy path. Physicians enter the virtual operating environment through VR devices, where they can view the following visual information: a 3D patient model (which can be freely rotated and scaled), the physiotherapy path planned by Artificial Intelligence (AI) (displayed as a semi-transparent green area), and the predicted electromagnetic field distribution (displayed as contour lines). Physicians can use VR controllers to grasp and adjust the control points of the physiotherapy path, modify physiotherapy parameters, such as setting the maximum temperature threshold to 42℃, and verify the rationality of the path from different perspectives.

[0080] AI planning is based on general models and algorithms, while physicians, understanding a patient's specific medical history and feelings, need to fine-tune the target area and energy intensity. Despite digital twins, slight differences may exist in the acoustic properties of individual tissues, requiring physicians to make predictive adjustments based on experience. The final approval of the treatment plan rests with the physician, and the adjustment process is a crucial step in fulfilling their medical responsibilities.

[0081] Verify the rationality of the treatment path. Physicians can freely rotate, zoom, and dissect the patient's digital twin model from multiple perspectives, observing the relationship between the planned physiotherapy path (usually highlighted) and the internal anatomical structures from any angle. View the energy field prediction; the system can overlay and display the simulated predicted sound intensity distribution cloud map. Physicians can intuitively determine whether the energy is focused on the target area and whether sensitive areas have been avoided. During the simulated physiotherapy process, animations of the robotic arm moving along the path can be played, allowing for the early detection of potential physical interferences (such as conflicts with patient positioning).

[0082] Therefore, adjustment instructions can include instructions for rotating, scaling, or slicing fused information; instructions for adjusting the target area of ​​the current physiotherapy pathway; and instructions for adjusting the energy intensity of the target area of ​​the current physiotherapy pathway. After the physician selects and approves the physiotherapy plan, an executable sequence of machine instructions is generated.

[0083] This embodiment provides an interactive interface for fine-tuning the physiotherapy path. Physicians can intuitively identify and manually avoid high-risk areas not recognized by the algorithm (such as paths close to nerve bundles) through VR adjustments, preventing iatrogenic injuries caused by mechanically executing the algorithm. Gestures for quickly cutting views, zooming in on lesions, and rotating the viewpoint significantly improve the efficiency and accuracy of fine-tuning while reducing cognitive load. This greatly enhances the patient's physiotherapy experience and compliance, preventing treatment interruptions due to discomfort.

[0084] In some embodiments, after the first adjustment module, the following is further included: The return module is used to take the adjusted current physiotherapy path as the new preset physiotherapy path and return to the preset physiotherapy target area obtained by identifying and processing the body surface temperature information for the next physiotherapy treatment, until the physiotherapy control ends when the preset duration is reached.

[0085] Specifically, during the execution of the current physiotherapy path, the actuator moves to a safe ready position and gradually approaches the area to be treated. The laser rangefinder is activated for precise positioning, and the dual treatment heads are positioned to form a capacitive field. Force sensors detect the contact pressure, and after confirming that there are no obstacles entering the treatment area, the system is ready.

[0086] After the current physiotherapy path is completed, the adjusted current physiotherapy path is used as the new preset physiotherapy path. The system then returns to the preset physiotherapy target area obtained by identifying and processing the body surface temperature information for the next physiotherapy treatment. This process continues until the preset physiotherapy duration is reached, at which point the physiotherapy control ends. Compared to open-loop control, this embodiment uses a closed-loop control method. Physiotherapy starts, ultrasonic waves are output, and the initial power is set to the AI ​​recommended value. The robotic arm initially moves along the preset physiotherapy path, and the cycle is monitored and adjusted in real time, gradually narrowing according to each physiotherapy path. Throughout the process, real-time data is read, safety is assessed, and efficacy is optimized and adjusted. Normal physiotherapy data is recorded. The code is as follows: while treatment_time <preset_time: #Read real-time data current_temp=thermal_camera.get_temperature() position=robot.get_current_position() #Safety Assessment if current_temp>safety_threshold: power_control.reduce_power(30%) #Power reduction protection robot.increase_speed(50%) # Speed ​​up the process log_safety_event(“Overheat protection activated”) #Efficacy Optimization and Adjustment elif current_temp <optimal_min: power_control.increase_power() #Normal physiotherapy range Else: maintain_current_parameters() #Data Recording data_logger.record(temperature=current_temp, position = position, power = current_power).

[0087] Furthermore, the anomaly handling mechanism is multi-layered and cannot rely solely on temperature. For physiotherapy anomalies (primarily based on temperature), such as slight overheating, an infrared thermal imager monitors the body surface temperature. If it exceeds a safety threshold, the system automatically reduces power or removes the therapy head. If the therapeutic effect is insufficient, and the temperature is below the treatment window, the system automatically increases power. For equipment anomalies (based on the equipment's own sensors), the robotic arm's encoder and ammeter can report abnormal joint movement or overload; the ultrasonic physiotherapy unit can monitor abnormal output power, circuit faults, and transducer impedance changes; force sensors can detect abnormal contact force between the therapy head and the body surface; and electronic fences or vision systems can detect external safety risks such as unauthorized entry. Once any of these sensors triggers an alarm, it can be determined as an equipment anomaly, initiating an emergency stop.

[0088] The closed-loop control process provided in this embodiment establishes a real-time control loop from temperature monitoring to AI decision-making and parameter adjustment, which enables temperature sampling and power adjustment every 200ms while reducing the risk of burns.

[0089] In some embodiments, it also includes: The second acquisition module is used to acquire fusion information after physiotherapy; The first analysis module is used to perform quantitative analysis based on the fusion information before and after physiotherapy to determine the information change indicators. The fourth generation module is used to generate an evaluation report based on the information change indicators.

[0090] Specifically, the physiotherapy termination and post-operative assessment follow a standard end procedure. Upon reaching the preset physiotherapy time, the power gradually decreases to zero, the robotic arm automatically returns to a safe position, and the system plays a physiotherapy end prompt. At this point, a rapid thermal imaging scan is performed to collect post-treatment temperature distribution data for subsequent quantitative efficacy analysis. This quantitative analysis uses objective data to replace subjective descriptions in assessing efficacy, making the physiotherapy measurable and comparable. Specific assessment methods are as follows: 1) Temperature change index: Compare infrared thermograms before and after physiotherapy to calculate the average temperature reduction in the target area and the proportion of reduction in the high-temperature area. 2) Energy distribution uniformity: Analyze the energy output and temperature data recorded during physiotherapy to calculate the uniformity coefficient of temperature distribution within the target area. The uniformity after physiotherapy should be better than before. 3) Dosage accuracy: Compare the expected dose planned by AI (based on simulated acoustic energy) with the cumulative dose actually output by the physiotherapy; the deviation should be less than a preset range (e.g., ±5%). 4) Report generation: After physiotherapy, the system automatically generates a graphic report, including comparative charts, providing physicians with objective efficacy assessment data and can be used for long-term medical record management and research analysis. The specific code is as follows: #Physiotherapy Effectiveness Evaluation Algorithm def evaluate_treatment(pretreatment_data, posttreatment_data): #Calculate temperature change delta_temp=calculate_temperature_change(pretreatment_data, posttreatment_data) #Computational uniformity improvement uniformity_improvement=calculate_uniformity_improvement() #Generate evaluation report report=generate_report(delta_temp, uniformity_improvement) return report Finally, the report is generated and the data is archived. An automatic physiotherapy report is produced, including a before-and-after temperature comparison chart, actual energy output curve, safety incident records, and a summary of physiotherapy parameters. The data is then archived, with the complete physiotherapy data stored in the patient's medical record, preparing the system for the next patient.

[0091] The physiotherapy report generation process provided in this embodiment automatically integrates multimodal data before and after physiotherapy (such as: temperature change curves in thermograms, point cloud deformation data, and energy deposition distribution maps). Through comparative charts before and after physiotherapy (such as: average temperature decrease in the lesion area, and blood circulation improvement index), abstract sensations are transformed into concrete numerical indicators. Physicians can use objective data to judge the effectiveness of the current physiotherapy session and decide whether to adjust the parameters (dosage, frequency, and pathway) for the next session, thus establishing a complete physiotherapy data traceability system.

[0092] Table 1 shows the corresponding physiotherapy process at key nodes. As shown in Table 1, within T+0 min, patient positioning and system initialization occur; within T+1 min, multimodal data is acquired synchronously; within T+3 min, the intelligent planning stage for multimodal data fusion and modeling takes place; within T+8 min, physician confirmation occurs; within T+12 min, physiotherapy is executed, and the robotic arm automatically positions itself; within T+27 min, physiotherapy is terminated and postoperative assessment is performed.

[0093] Table 1. Corresponding physiotherapy processes under key nodes

[0094] Furthermore, this application also provides an ultrasonic physiotherapy device, including a controller, an execution component, and a treatment head; wherein, the execution component is a robotic arm, and the treatment head includes a sensing module; the sensing module is connected to the controller; the controller is connected to the robotic arm; and the robotic arm is connected to the treatment head; Controller, including the control device for ultrasonic physiotherapy instruments.

[0095] Figure 2 A structural diagram of an ultrasonic physiotherapy device provided in an embodiment of this application is shown below. Figure 2 As shown, the sensing module collects spatial geometric information, surface temperature information, and image information of the area to be treated, and sends them to the controller 16 for mapping processing to obtain fused information. Based on the fused information and corresponding constraints, a preset treatment path is determined. Within the preset treatment path, the surface temperature information is identified to obtain a preset treatment target area. The current treatment path is determined based on the preset treatment target area and the fused information. The path parameters of the current treatment path are adjusted to obtain the adjusted current treatment path. The controller 16 executes the processing through an execution component, which is the robotic arm 17, to perform the subsequent treatment path execution. Its end is the treatment head 15 for treatment.

[0096] For a description of the control device for an ultrasonic physiotherapy device provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the ultrasonic physiotherapy device described above.

[0097] In some embodiments, the controller 16 is equipped with a VR interaction interface; Connect VR device 18 via VR interactive interface; The controller 16 is used to receive adjustment instructions; and to adjust the path parameters of the current physiotherapy path according to the adjustment instructions to obtain the adjusted current physiotherapy path.

[0098] Specifically, the adjustment instructions are received from the VR device 18 via the VR interaction interface. For the VR device 18, the subsequent adjustment instructions are issued based on the physician wearing it in the VR environment, in order to adjust the path parameters of the current physiotherapy path to obtain the adjusted current physiotherapy path. Specific adjustment instructions can be found in the above embodiments and will not be elaborated upon here.

[0099] This embodiment allows physicians to directly control the host computer via VR devices to fine-tune automatically generated physiotherapy paths, upgrading human-computer interaction from abstract operations on a two-dimensional screen to immersive and intuitive operations in three-dimensional space. The stereoscopic vision provided by VR allows doctors to clearly determine the occlusion relationship and distance between the path and bones, nerves, and blood vessels, completely eliminating depth ambiguity from a planar perspective. This significantly reduces cognitive load and improves operational accuracy. It also greatly enhances obstacle avoidance precision, ensuring energy is precisely applied to the target area while minimizing the avoidance of sensitive tissues, achieving truly millimeter-level safe control.

[0100] Furthermore, this application also provides a control method for an ultrasonic physiotherapy device. Figure 3 A flowchart of a control method for an ultrasonic physiotherapy device provided in this application embodiment is shown below. Figure 3 As shown, it includes: S11: Collect spatial geometric information, body surface temperature information, and image information of the area to be treated, and perform mapping processing to obtain fused information; S12: Determine the preset physiotherapy path based on the fused information and corresponding constraints; S13: Within the preset physiotherapy path, the body surface temperature information is identified and processed to obtain the preset physiotherapy target area; and the current physiotherapy path is determined based on the preset physiotherapy target area and fusion information; S14: Adjust the path parameters of the current physiotherapy path to obtain the adjusted current physiotherapy path, so that the execution component can complete the current control process.

[0101] For a description of the control method for an ultrasonic physiotherapy device provided in this application, please refer to the above-described method embodiments. This application will not repeat the description here, as it has the same beneficial effects as the control device for the ultrasonic physiotherapy device described above.

[0102] Figure 4 A structural diagram of the control device for another ultrasonic physiotherapy device provided in the embodiments of this application is shown below. Figure 4 As shown, the device includes: Memory 21 is used to store computer programs; Processor 22 is used to execute computer programs to implement the steps of the control method for the ultrasonic physiotherapy device.

[0103] The control device for the ultrasonic physiotherapy device provided in this embodiment may include, but is not limited to, smartphones, tablets, laptops, or desktop computers.

[0104] The processor 22 may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor 22 may be implemented using at least one of the following hardware forms: Digital Signal Processor (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 22 may also include a main processor and a coprocessor. The main processor, also known as the Central Processing Unit (CPU), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor 22 may integrate a Graphics Processing Unit (GPU), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor 22 may also include an AI processor, which handles computational operations related to machine learning.

[0105] The memory 21 may include one or more computer-readable storage media, which may be non-transitory. The memory 21 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In this embodiment, the memory 21 is used to store at least the following computer program 211, which, after being loaded and executed by the processor 22, is capable of implementing the relevant steps of the control method of the ultrasonic physiotherapy device disclosed in any of the foregoing embodiments. In addition, the resources stored in the memory 21 may also include an operating system 212 and data 213, etc., and the storage method may be temporary storage or permanent storage. The operating system 212 may include Windows, Unix, Linux, etc. The data 213 may include, but is not limited to, the data involved in the control method of the ultrasonic physiotherapy device.

[0106] In some embodiments, the control device of the ultrasonic physiotherapy instrument may further include a display screen 23, an input / output interface 24, a communication interface 25, a power supply 26, and a communication bus 27.

[0107] Those skilled in the field can understand, Figure 4 The structure shown does not constitute a limitation on the control device of the ultrasonic physiotherapy device and may include more or fewer components than shown.

[0108] The processor 22 implements the control method of the ultrasonic physiotherapy device provided in any of the above embodiments by calling the instructions stored in the memory 21.

[0109] Figure 5 A flowchart illustrating another control method for an ultrasonic physiotherapy device provided in this application embodiment is shown below. Figure 5 As shown, it includes: Phase 1: Multimodal data fusion and data twin model construction: S21: Collect impact data; S22: Acquire 3D visual and infrared thermal imaging data; S23: Multimodal data registration and fusion; S24: Construct a digital twin model with physical attributes; Phase Two: Simulation-Based Pre-Therapy Planning S25: Define the physical therapy target area; S26: Ultrasonic sound field simulation; S27: Key indicators for calculating sound intensity and temperature field distribution; S28: The optimization algorithm searches for the optimal solution to generate the optimal solution; S29: Generate individualized physiotherapy plans; S30: Determine whether the doctor has confirmed through VR device; if yes, proceed to step S31; otherwise, proceed to step S32. S32: Modify the path parameters of the physiotherapy path corresponding to the physiotherapy plan, and return to step S26; Phase Three: Implementation and Closed-Loop Control of Physiotherapy; S31: The robotic arm precisely executes the physiotherapy plan; S33: Real-time monitoring and feedback; S34: Determine whether the safety and efficacy standards are met. If yes, proceed to step S35; otherwise, proceed to step S36. S35: Physiotherapy ends; S36: Dynamically optimize the control parameters and return to step S31.

[0110] For a description of the control device for an ultrasonic physiotherapy device provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the control method for the ultrasonic physiotherapy device described above.

[0111] Furthermore, this application also provides a computer-readable storage medium storing a computer program, which, when executed by processor 22, implements the steps of the control method for the ultrasonic physiotherapy device described above.

[0112] It is understood that if the methods in the above embodiments are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and executes all or part of the steps of the methods in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0113] For a description of the computer-readable storage medium provided in this application, please refer to the above method embodiments. This application will not repeat the description here, but it has the same beneficial effects as the control method of the ultrasonic physiotherapy device described above.

[0114] The control device for an ultrasonic physiotherapy instrument and the ultrasonic physiotherapy instrument provided in this application have been described in detail above. The various embodiments in the specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the device disclosed in the embodiments, since it corresponds to the method disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make several improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of this application.

[0115] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes the element.

Claims

1. A control device for an ultrasonic physiotherapy instrument, characterized in that, include: The first acquisition module is used to acquire spatial geometric information, body surface temperature information and image information of the area to be treated, and perform mapping processing to obtain fused information; The first determining module is used to determine a preset physiotherapy path based on the fusion information and the corresponding constraints. The second determining module is used to identify and process the body surface temperature information within the preset physiotherapy path to obtain the preset physiotherapy target area; The current physiotherapy path is determined based on the preset physiotherapy target area and the fusion information; The first adjustment module is used to adjust the path parameters of the current physiotherapy path to obtain the adjusted current physiotherapy path, so that the execution component can perform the processing to complete the current control processing.

2. The control device for the ultrasonic physiotherapy instrument according to claim 1, characterized in that, The first acquisition module includes: The first generation module is used to scan the area to be treated using a three-dimensional scanning method to generate three-dimensional point cloud information to form a physiotherapy reference coordinate system; wherein, the three-dimensional point cloud information includes the sensitive areas corresponding to the bone tissue structure and key risk anatomical structures of the area to be treated. The second generation module is used to generate a temperature data heat map by scanning a preset area corresponding to the area to be treated with an infrared thermal imager, and attach it to the position corresponding to the three-dimensional point cloud information. The first extraction module is used to extract feature information from the attached 3D point cloud information and the image information using an iterative nearest-point algorithm, and to iteratively output a rotation matrix; The first transformation module is used to transform the image information into the physiotherapy reference coordinate system according to the rotation matrix to obtain the fused information.

3. The control device for the ultrasonic physiotherapy instrument according to claim 2, characterized in that, The first determination module includes: The first output module is used to pre-call the pre-trained model and input the image information of the fused information to output a set of key point coordinates; The third determining module is used to determine the theoretical coordinate information of the target acupoint based on the bone measurement method and the set of key point coordinates, so as to determine the initial acupoint candidate list. The first correction processing module is used to correct the preset target coordinates based on the body surface temperature information of the initial acupoint candidate list and the fused information to obtain the corrected target coordinates. The fourth determining module is used to determine the preset physiotherapy path based on the target coordinates and the constraints.

4. The control device for the ultrasonic physiotherapy instrument according to claim 3, characterized in that, The fourth determination module includes: The first acquisition module is used to acquire the sensitive region corresponding to the fused information; The fifth determining module is used to determine the first straight-line path from the entrance of the ultrasonic physiotherapy head to the target point coordinates; The first module is used to designate the intersecting area as a risk area when the first straight path intersects with the sensitive area. The sixth determining module is used to determine a first preset area, centered on the risk area, as an obstacle avoidance constraint area; The seventh determining module is used to determine the depth information between the entrance and the target point coordinates; The eighth determining module is used to determine the preset physiotherapy path based on the relationship between the obstacle avoidance constraint area and the risk area, and the relationship between the depth information and the depth constraint information.

5. The control device for the ultrasonic physiotherapy instrument according to claim 4, characterized in that, The second determining module includes: The second extraction module is used to extract potential physiotherapy target areas corresponding to body surface temperature information that are higher than preset body surface temperature information from the body surface temperature information. The first screening module is used to screen out potential physiotherapy target areas corresponding to unobstructed sound beam paths from the entrance to the potential physiotherapy target area, so as to use them as the first target physiotherapy target areas. The second screening module is used to screen out a second target physiotherapy target area that is at a preset distance from the sensitive area within the first target physiotherapy target area, and to use the second target physiotherapy target area as the preset physiotherapy target area.

6. The control device for the ultrasonic physiotherapy instrument according to claim 5, characterized in that, The second determining module includes: The third generation module is used to generate an ultrasonic sound field distribution based on different preset distances between the preset physiotherapy target area and the sensitive area. The ninth determining module is used to determine the target electrode plate spacing based on the location information, depth information, and relative position information of the preset physiotherapy target area and the sensitive area in the ultrasonic sound field distribution. The tenth determining module is used to calculate the surface acoustic intensity in reverse based on the depth information and the tissue average attenuation coefficient, so as to determine the output power to be achieved. The eleventh determining module is used to determine the current physiotherapy path based on the shape of the preset physiotherapy target area, the spacing between the target electrode plates, and the output power.

7. The control device for the ultrasonic physiotherapy instrument according to claim 1, characterized in that, The first adjustment module includes: The first receiving module is used to receive adjustment instructions through an interactive interface; The second adjustment module is used to adjust the path parameters of the current physiotherapy path according to the adjustment instruction to obtain the adjusted current physiotherapy path; The adjustment instruction is one of at least one of the following: Instructions for rotating, scaling, and slicing the fused information accordingly; An instruction to adjust the target area of ​​the current physiotherapy pathway; An instruction to adjust the energy intensity of the target area of ​​the current physiotherapy pathway.

8. The control device for the ultrasonic physiotherapy instrument according to claim 1, characterized in that, Following the first adjustment module, it also includes: The return module is used to take the adjusted current physiotherapy path as the new preset physiotherapy path and return to the preset physiotherapy target area obtained by identifying and processing the body surface temperature information, so as to perform the next physiotherapy treatment until the physiotherapy duration reaches the preset duration and the physiotherapy control ends.

9. The control device for the ultrasonic physiotherapy instrument according to claim 8, characterized in that, Also includes: The second acquisition module is used to acquire the fusion information after physiotherapy; The first analysis module is used to perform quantitative analysis based on the fusion information before and after physiotherapy to determine the information change indicators. The fourth generation module is used to generate an evaluation report based on the information change indicators.

10. An ultrasonic physiotherapy device, characterized in that, It includes a controller, an actuator, and a treatment head; wherein the actuator is a robotic arm, and the treatment head includes a sensing module; the sensing module is connected to the controller; the controller is connected to the robotic arm; and the robotic arm is connected to the treatment head. The controller includes the control device of the ultrasonic physiotherapy device according to any one of claims 1 to 9.

Images