Radiant extracorporeal shock wave protocol method and system responsive to swallowing training data

Abstract

The present application relates to the technical field of swallowing rehabilitation treatment, in particular to a radial extracorporeal shock scheme regulation method and system responding to swallowing training data, comprising: arranging a sensor array on a target point area of a training person's body surface to capture multi-channel physiological signals of standard swallowing actions, extracting time sequence, force distribution and coordination biological mechanics parameters through pattern analysis. The parameters are input into a swallowing function evaluation atlas library for comparison to generate a quantitative function deviation index, and based on the index, a multidimensional space search is performed on a basic extracorporeal shock scheme composed of shock energy, pulse sequence and action angle. The real-time change trend of force distribution and coordination is combined to fine-tune the basic scheme parameters to generate customized extracorporeal shock treatment instructions. The method realizes quantitative analysis of swallowing mechanical characteristics and adaptive customization of shock scheme, and makes the extracorporeal shock treatment fit the real-time state of individual swallowing function.

Description

Technical Field

[0001] This invention relates to the field of swallowing rehabilitation therapy technology, and in particular to a method and system for regulating a radial extracorporeal shock therapy program in response to swallowing training data. Background Technology

[0002] Current rehabilitation treatments for dysphagia often employ external invasive percussion devices (EIPs), with device parameters frequently set manually based on clinical experience. Swallowing movement monitoring relies on a single type of sensor to collect local physiological signals, resulting in a limited range of signal dimensions. Swallowing function assessments primarily rely on imaging examinations or subjective scales, with results mainly being qualitative descriptions. The industry lacks a standardized swallowing movement pattern database encompassing normal states to various dysphagia states, and the methods for analyzing the correlation between physiological signals and swallowing biomechanical characteristics are rather rudimentary.

[0003] Current extracorporeal shock therapy protocols mostly employ fixed parameters, making it impossible to dynamically adjust the shock energy, pulse sequence, and angle of action based on swallowing function. Protocol selection also lacks quantitative reference. The temporal, force distribution, and coordination characteristics of swallowing movements cannot be fully extracted and analyzed in real time. Adjustments to shock parameters lack corresponding real-time biomechanical data to support them, resulting in a low degree of compatibility between extracorporeal shock therapy protocols and the individual swallowing function of trainees.

[0004] It is impossible to extract swallowing timing, force distribution and coordination-related biomechanical parameters through multi-channel physiological signal analysis and generate quantitative functional deviation indicators. It is impossible to complete the adaptive retrieval of basic protocols in the multi-dimensional impact protocol space composed of impact energy, pulse sequence and action angle. It is impossible to make parameterized fine-tuning of the impact protocol based on the real-time changing trend of swallowing force distribution and coordination. It is difficult to generate customized extracorporeal impact therapy instructions that fit the individual's real-time state. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of the prior art and to propose a method and system for regulating a radial extracorporeal shock scheme in response to swallowing training data.

[0006] To achieve the above objectives, the present invention employs the following technical solution: a method for regulating a radial extracorporeal shock protocol in response to swallowing training data, comprising:

[0007] A sensor array is arranged in a pre-defined stimulation target area on the trainee's body surface. The sensor array captures multi-channel physiological signals generated when the trainee performs a standard swallowing action. Pattern analysis was performed on the multi-channel physiological signals to extract a series of biomechanical parameters characterizing the swallowing mechanics. These biomechanical parameters include the temporal characteristics, force distribution characteristics, and coordination characteristics of the swallowing action. The biomechanical parameters are input into a swallowing function assessment atlas for comparison. The swallowing function assessment atlas stores standard swallowing patterns from normal to various functional impairment states. Based on the comparison results, a quantitative functional deviation index is generated, which is used to quantify the difference between the current swallowing action and the target pattern; Based on the functional deviation index, an adaptive search is performed in a multi-dimensional impact scheme space to select a basic radial extracorporeal impact scheme. The multi-dimensional impact scheme space consists of three dimensions: impact energy, pulse sequence, and action angle. The selected basic extracorporeal shock therapy scheme is parametrically fine-tuned based on the real-time changing trends of the force distribution and coordination characteristics in the biomechanical parameters, thereby generating a customized extracorporeal shock therapy instruction.

[0008] As a further aspect of the present invention, pattern analysis is performed on the multi-channel physiological signals to extract a series of biomechanical parameters characterizing swallowing mechanics, including: The raw physiological signals from the sensor array are preprocessed to remove power frequency interference and motion artifacts; Blind source separation was performed on the preprocessed signal to separate independent signal components directly related to hyoid bone elevation, larynx forward movement, and pharyngeal contraction. Envelope extraction and peak detection were performed on each independent signal component to determine the start time, peak time and duration of activation of each swallowing-related muscle group; Calculate the time difference between the peak times of different independent signal components as a temporal indicator of muscle coordination; By integrating the envelope area and peak amplitude of each independent signal component, a multidimensional force distribution vector is generated, which is the force distribution feature.

[0009] As a further aspect of the present invention, the biomechanical parameters are input into a swallowing function assessment atlas library for comparison, including: A current swallowing feature vector is constructed from the biomechanical parameters, the current swallowing feature vector including the temporal features, force distribution features and coordination features; The current swallowing feature vector is matched with a predefined set of reference feature vectors in the swallowing function assessment atlas library; Each vector in the reference feature vector set is associated with a specific swallowing dysfunction pattern and is labeled with an ideal correction target pattern; Select the reference feature vector that has the highest similarity to the current swallowing feature vector, and determine the associated functional impairment pattern as the current primary functional impairment pattern; Calculate the Euclidean distance between the current swallowing feature vector and the ideal correction target pattern vector corresponding to the main functional impairment pattern, and use the normalized Euclidean distance as the functional deviation index.

[0010] As a further aspect of the present invention, the construction process of the swallowing function assessment atlas library includes: High-precision physiological signals and synchronous video fluorescence imaging data were collected from a large number of subjects under different swallowing tasks; Based on the video fluorescence angiography data, experts will perform functional scoring and pattern labeling for each swallowing action to establish a real functional status label; Standardized biomechanical parameters are extracted from the high-precision physiological signals to form sample feature vectors that correspond one-to-one with the functional state labels; Cluster analysis was used to classify the feature vectors of the samples, forming different swallowing action pattern clusters. The central feature vector of each pattern cluster is used as the reference feature vector, and its corresponding typical functional state is used as the label of the vector. Both are stored in the swallowing function assessment atlas library.

[0011] As a further aspect of the present invention, based on the functional deviation index, an adaptive search is performed in a multidimensional impact scheme space to select a basic extracorporeal radiation impaction scheme, including: A three-dimensional orthogonal coordinate system is defined as the space of the multi-dimensional impact scheme, with the three coordinate axes representing the impact energy level, pulse sequence mode encoding, and impact angle encoding, respectively. In the multidimensional impact scheme space, a series of discrete nodes are pre-calibrated, and each node corresponds to a verified and effective basic radial extracorporeal impact scheme. The functional deviation index is mapped to a search radius in the multidimensional impact scheme space; Using the historical mapping point corresponding to the current swallowing action mode in the multidimensional impact scheme space as the center of the sphere and the search radius as the radius, a candidate scheme sphere is defined. From all discrete nodes located within the candidate scheme sphere, select the node with the highest historical correction success rate of the main functional impairment pattern, and use the impact scheme corresponding to that node as the selected basic radial extracorporeal impact scheme.

[0012] As a further aspect of the present invention, the selected basic extracorporeal shock wave therapy scheme is parametrically fine-tuned to generate a customized extracorporeal shock wave therapy instruction, including: The basic radial extracorporeal impact scheme was analyzed to obtain its initial impact energy, standard pulse sequence, and reference action angle. A real-time feedback controller is established, the input of which is the real-time changing trend of the force distribution characteristics and coordination characteristics; The real-time feedback controller calculates and generates a set of parameter fine-tuning values ​​based on the changing trend of the input. The parameter fine-tuning values ​​include energy compensation value, pulse interval adjustment amount and angle offset amount. The initial impact energy, standard pulse sequence, and reference action angle are superimposed with the corresponding parameter fine-tuning amounts to obtain the fine-tuned impact parameters; Based on the fine-tuned impact parameters, a customized extracorporeal shock therapy instruction that can be executed by the machine is generated, which specifies in detail the waveform, timing and spatial orientation of the energy output.

[0013] As a further aspect of the present invention, it also includes: The customized extracorporeal shock therapy command is sent to a radial extracorporeal shock generator; The radial extracorporeal shock generator, according to the customized extracorporeal shock therapy command, drives its energy output array to apply focused mechanical shock to the stimulation target area. While the impact is being applied, the sensor array simultaneously collects the trainee's real-time response physiological signals; Based on the real-time response physiological signals, the functional deviation index is dynamically updated, and based on the updated functional deviation index, the next round of adjustment to the customized extracorporeal shock therapy command is initiated. The radial extracorporeal shock generator, according to the customized extracorporeal shock therapy command, drives its energy output array to apply focused mechanical shock to the stimulation target area, including: The central controller of the radial extracorporeal shock generator analyzes the customized extracorporeal shock therapy command and decomposes it into independent drive signals for each transducer unit in the energy output array. Each transducer unit generates an ultrasonic pulse of a specific frequency and phase according to its corresponding independent drive signal; By controlling the phase difference of the ultrasonic pulses emitted by each transducer unit, constructive interference is formed at the spatial position of the stimulation target area, thereby generating a high-energy-density mechanical impact focus. The mechanical impact focus applies periodic or modulated mechanical loading to the inside or surface of the stimulation target area according to the pulse sequence specified in the instruction.

[0014] As a further aspect of the present invention, while applying the impact, the sensor array simultaneously acquires the trainee's real-time response physiological signals, including: During the interval of the output impact pulse from the radial external impact generator or at a specific synchronization moment, the sensor array is triggered to perform high-speed sampling; Collect induced electromyographic signals, tissue vibration signals, and residual physiological signals of voluntary swallowing movements generated by extracorporeal impact intervention; The acquired signals are labeled as the real-time response physiological signals and timestamped with the currently executing shock therapy command. Based on the real-time response physiological signals, the functional deviation index is dynamically updated, including: Pattern analysis was performed on the real-time response physiological signals to extract a new set of biomechanical parameters after the impact intervention; Calculate the instantaneous change vector between the new biomechanical parameters after the impact intervention and the original biomechanical parameters before the impact intervention. Project the instantaneous change vector onto the direction of the ideal correction target pattern vector, and calculate the length and direction of the projection; Based on the length and direction of the projection, the value of the functional deviation index is corrected. If the projection direction is consistent with the desired correction direction and the length is significant, the value of the functional deviation index is reduced; otherwise, corresponding adjustments are made to obtain the updated functional deviation index.

[0015] As a further aspect of the present invention, based on the updated functional deviation index, the next round of adjustments to the customized extracorporeal shock therapy instructions is initiated, including: The updated functional deviation index is input into the adaptive retrieval process, but the search center is updated to the node corresponding to the currently executed impact scheme in the multidimensional impact scheme space. Re-execute the adaptive search to obtain a new basic extracorporeal radiological shock protocol or confirm that the original protocol is maintained. If a new basic extracorporeal shock therapy protocol is obtained, a new customized extracorporeal shock therapy instruction will be generated by combining the latest real-time trends and through a parameterized fine-tuning process. In the next treatment cycle or at a preset adjustment time, a new customized extracorporeal shock therapy command is sent to the radial extracorporeal shock generator to replace or modify the command being executed.

[0016] As a further aspect of the present invention, the present invention also includes a radial extracorporeal shock protocol control system in response to swallowing training data. The system includes a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, it implements the steps of the radial extracorporeal shock protocol control method in response to swallowing training data as described above.

[0017] Compared with the prior art, the advantages and positive effects of the present invention are as follows: A sensor array is deployed at pre-defined stimulation target areas on the trainee's body surface to capture multi-channel physiological signals of swallowing movements. Pattern analysis of these multi-channel physiological signals allows for the extraction of biomechanical parameters corresponding to the temporal characteristics, force distribution characteristics, and coordination characteristics of swallowing movements. These parameters are then input into an assessment atlas library of standard patterns for normal and various swallowing dysfunctions for comparison. This directly generates a functional deviation index that quantifies the difference between swallowing movements and the target pattern. Multi-channel signal acquisition can fully reflect the overall mechanical performance of swallowing movements, and standardized atlas library comparison can eliminate judgment bias caused by subjective assessment. The functional deviation index can directly present the quantitative difference in swallowing function.

[0018] By adaptively searching within a multidimensional impact protocol space constructed from impact energy, pulse sequence, and angle of action based on the functional deviation index, a basic radial external impact protocol matching the current swallowing function can be quickly selected. The basic protocol is then fine-tuned parametrically by combining the real-time changing trends of force distribution and coordination characteristics in biomechanical parameters. This allows the various parameters of the impact protocol to align with the trainee's real-time swallowing mechanics. The basis for parameter fine-tuning is directly derived from the real-time mechanical changes in swallowing movements. The resulting customized external impact therapy instructions ensure that the form of external impact action is consistent with the real-time changes in the individual trainee's swallowing function, and that the execution logic of the impact protocol matches the dynamic changes in swallowing physiological movements. Attached Figure Description

[0019] Figure 1 This is a flowchart of the radial extracorporeal shock protocol control method in response to swallowing training data as described in this invention; Figure 2 A flowchart for comparing swallowing function assessment atlases and calculating functional deviation. Figure 3 A graph showing the changing trend of force distribution characteristics during swallowing; Figure 4 Normalized time series diagram of physiological signals of swallowing muscle groups; Figure 5 This is a time-series diagram of the physiological response to swallowing under radial extracorporeal shock intervention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0021] In the description of this invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, in the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0022] See Figure 1 A sensor array is deployed at pre-defined stimulation target areas on the trainee's body surface to capture multi-channel physiological signals generated during standard swallowing movements. Pattern analysis is performed on the captured multi-channel physiological signals to extract a series of biomechanical parameters characterizing swallowing mechanics, including the temporal, force distribution, and coordination features of the swallowing movements. These extracted biomechanical parameters are then compared against a pre-built swallowing function assessment atlas, which stores standard swallowing patterns ranging from normal to various functional impairment states. Based on the comparison results, a quantitative functional deviation index is generated to quantify the difference between the current swallowing movement and the target pattern. Based on the calculated functional deviation index, an adaptive search is performed in a multi-dimensional impact scheme space consisting of impact energy, pulse sequence, and action angle to select a basic radial extracorporeal impact scheme. The selected basic radial extracorporeal impact scheme is then fine-tuned parametrically based on the real-time trends of force distribution and coordination features in the biomechanical parameters, ultimately generating a customized extracorporeal impact therapy instruction.

[0023] In one embodiment of the invention, the sensor array includes multiple high-precision surface electromyography (EMG) sensors and accelerometers. The sensor array captures multi-channel physiological signals generated when the trainee performs a standard swallowing action. The multi-channel physiological signals include multiple raw EMG signals and vibration signals. The raw physiological signals from the sensor array are transmitted to a signal processing unit for preprocessing. The preprocessing steps include applying a band-stop filter to remove power frequency interference and using an adaptive filtering algorithm to eliminate motion artifacts caused by the trainee's head micro-movements. Blind source separation is performed on the preprocessed signals. In some embodiments, an independent component analysis algorithm is used to decompose the mixed signal and separate three independent signal components directly related to hyoid bone elevation, larynx forward movement, and pharyngeal contraction. Envelope extraction and peak detection are performed on each independent signal component. Envelope extraction uses the Hilbert transform method, and peak detection is achieved by finding points where the first derivative of the signal is zero and the second derivative is negative, thereby determining the start time, peak time, and duration of activation of each swallowing-related muscle group. The time difference between the peak times of different independent signal components is calculated, such as the difference between the peak time of the hyoid bone elevation component and the peak time of the pharyngeal contraction component. This time difference serves as a temporal indicator of muscle coordination. The envelope area and peak amplitude of each independent signal component are integrated. The envelope area is calculated through numerical integration, and the peak amplitude is obtained directly from the detection results to generate a three-dimensional force distribution vector. Each dimension of the force distribution vector corresponds to the mechanical contribution of an independent signal component, and the force distribution vector is the force distribution feature.

[0024] In specific implementation, please refer to Figure 2 A current swallowing feature vector is constructed from biomechanical parameters. This vector includes temporal features, force distribution features, and coordination features. The temporal features include the onset, peak, and duration of activation for each muscle group. The force distribution features are the aforementioned three-dimensional force distribution vector, and the coordination features are the temporal indices of muscle coordination. The current swallowing feature vector is then matched with a predefined set of reference feature vectors in a swallowing function assessment atlas. Each vector in the reference feature vector set is associated with a swallowing dysfunction pattern and is labeled with an ideal correction target pattern. The matching process is completed by calculating the Euclidean distance or cosine similarity between the vectors. The reference feature vector with the highest similarity to the current swallowing feature vector is selected, and its associated functional dysfunction pattern is determined as the current primary functional dysfunction pattern, for example, "delayed hyoid elevation with weak pharyngeal contraction." The Euclidean distance between the current swallowing feature vector and the ideal correction target pattern vector corresponding to the primary functional dysfunction pattern is calculated, and this normalized Euclidean distance is used as a functional deviation index. It is understandable that the normalization process maps the Euclidean distance to the interval between 0 and 1, and the value of the functional deviation index directly quantifies the severity of functional impairment.

[0025] In a practical implementation, an example scenario involves assessing a subject with delayed pharyngeal initiation. A sensor array collects physiological signals during a 5 ml water swallowing task. After preprocessing and blind source separation, the signals yield three independent signal components characterizing hyoid elevation, laryngeal anterior displacement, and pharyngeal contraction. Peak detection shows that the peak time of the pharyngeal contraction component is delayed by 220 milliseconds compared to the peak time of the hyoid elevation component. The calculated force distribution vector is [0.85, 1.20, 0.60], where the amplitude of the third component (corresponding to pharyngeal contraction) is relatively low. The constructed current swallowing feature vector includes the aforementioned temporal and vector data. After matching this vector with a swallowing function assessment atlas library, a reference feature vector for "delayed pharyngeal contraction initiation with insufficient force" is found, with a similarity of 92%. In the ideal correction target pattern vector associated with this reference feature vector, the peak time delay of the corresponding pharyngeal contraction component should be less than 100 milliseconds, and the target amplitude of the third component of the force distribution vector should be 1.10. The Euclidean distance between the current feature vector and the target vector is calculated and normalized, resulting in a function deviation index of 0.65. Optionally, the function deviation index can be calculated using the following formula:

[0026] in: As a functional deviation index, This is the current swallowing feature vector. For the ideal correction target pattern vector, Describes the Euclidean norm of a vector. This is the preset maximum normalization coefficient.

[0027] In one embodiment of the present invention, the construction of the swallowing function assessment atlas library begins with the data acquisition phase. A large number of subjects are collected using high-precision physiological signals and simultaneous video-fluorescence imaging data under different swallowing tasks. The high-precision physiological signals are recorded by a multi-channel sensor array placed in the neck, and the simultaneous video-fluorescence imaging data is acquired using X-ray fluoroscopy. Clinical experts use the video-fluorescence imaging data to perform functional scoring and pattern labeling for each swallowing action. Functional scoring can use standardized scales such as the osmosis-aspiration scale. Pattern labeling includes descriptions of items such as the amplitude of tongue retraction, the timing of laryngeal closure, and the integrity of pharyngeal peristaltic waves, thereby establishing realistic functional state labels. Standardized biomechanical parameters, including timing, force, and coordination characteristics, are extracted from the high-precision physiological signals to form sample feature vectors that correspond one-to-one with the functional state labels. Cluster analysis is used to classify the sample feature vectors. In some embodiments, the K-means clustering algorithm is used to analyze more than two thousand sample feature vectors, setting the number of clusters to five, forming five different swallowing action pattern clusters. The central feature vector of each pattern cluster is used as a reference feature vector, and its corresponding typical functional state is used as the label of the reference feature vector. Both are stored together in the swallowing function assessment atlas library, which is stored in a database format and supports high-speed vector similarity retrieval. In essence, the establishment of the swallowing function assessment atlas library ensures that any current swallowing feature vector can find its closest pathophysiological pattern classification within the library.

[0028] In practical implementation, a multidimensional impact protocol space is defined. This space is a three-dimensional orthogonal coordinate system, with the three axes representing the impact energy level, pulse sequence pattern encoding, and action angle encoding, respectively. Within this space, a series of discrete nodes are pre-labeled. Each discrete node corresponds to a validated basic radial extracorporeal shock therapy protocol, and the coordinates of these nodes are determined through extensive prior clinical trials. Functional deviation indices are mapped to a search radius within the multidimensional impact protocol space. This mapping ensures that larger functional deviation indices correspond to smaller search radii, enabling precise and conservative intervention for severe functional impairments. A candidate protocol sphere is defined with the historical mapping point corresponding to the current swallowing pattern in the multidimensional impact protocol space as its center and the search radius as its radius. The historical mapping point is obtained by querying a historical treatment record database. From all discrete nodes within the candidate protocol sphere, the node with the highest historical correction success rate for the primary functional impairment pattern is selected. This historical correction success rate is statistically derived from the treatment record database. The impact protocol corresponding to this discrete node is then chosen as the basic radial extracorporeal shock therapy protocol. Optionally, the search radius can be calculated using a linear or nonlinear mapping based on the functional deviation index. One feasible mapping formula is as follows:

[0029] in: Indicates the search radius. Indicators representing functional deviation. It is a scaling factor used to control the baseline size of the search radius. The exponential coefficient is used to adjust the sensitivity of the mapping. The formula ensures that the higher the functional deviation index, the smaller the search radius and the more concentrated the retrieval range.

[0030] In a specific example scenario and data comparison, the construction of the swallowing function assessment atlas library involved collecting data from 200 subjects, including healthy individuals and patients with different swallowing disorders, resulting in a total of 3000 effective swallowing sample feature vectors. After cluster analysis, the sample feature vectors were divided into five main swallowing action pattern clusters. For example, the central feature vector of cluster one was labeled "normal swallowing," cluster two was labeled "insufficient tongue thrust," cluster three was labeled "delayed pharyngeal initiation," cluster four was labeled "incomplete laryngeal elevation," and cluster five was labeled "multi-stage coordination disorder." The central feature vectors and labels of these five clusters constitute the core of the atlas library. When a "delayed pharyngeal initiation" pattern sample with a functional deviation index of 0.72 after example analysis is received, the system first determines its historical mapping point coordinates. The historical mapping point may be energy level 5, pulse code 3, or angle code 8. Substituting the functional deviation index of 0.72 into the formula, setting the proportional coefficient α to 10 and the exponential coefficient β to 2, the search radius R is calculated to be approximately 0.784. Within a candidate sphere centered at the historical mapping point (5,3,8) with a radius of 0.784, there are three discrete nodes, each corresponding to one of three different basic radial extracorporeal shock wave (ECW) protocols. A query of the historical database shows that these three protocols have historical correction success rates of 65%, 78%, and 82% for the "delayed pharyngeal initiation" pattern, respectively. The system automatically selects the basic ECW protocol corresponding to the discrete node with a historical correction success rate of 82%.

[0031] In one embodiment of the present invention, a basic radial extracorporeal shock wave (LES) protocol is analyzed. This protocol is stored in a data structure. The analysis operation reads fields from the data structure to obtain the specific values ​​of the initial shock energy, standard pulse sequence, and reference angle. The initial shock energy is expressed in joules. The standard pulse sequence is defined by a series of pulse widths and pulse intervals. The reference angle is represented by azimuth and elevation angles in the three-dimensional treatment head coordinate system. A real-time feedback controller is established. The input to the real-time feedback controller is the real-time trend of force distribution characteristics and coordination characteristics. This real-time trend is obtained by calculating the rate of change of each dimension of the force distribution vector and the temporal index of muscle coordination in the most recent swallowing movements. Based on the input trend, the real-time feedback controller calculates and generates a set of parameter fine-tuning values. In some embodiments, the real-time feedback controller is a proportional-integral controller. The parameter fine-tuning values ​​include energy compensation values, pulse interval adjustment values, and angle offset values. The unit of energy compensation values ​​is joules, the unit of pulse interval adjustment values ​​is milliseconds, and the unit of angle offset values ​​is degrees. It can be understood that the core function of the real-time feedback controller is to translate the dynamic changes of physiological signals into adjustment instructions for treatment parameters.

[0032] In practical implementation, the calculation of parameter fine-tuning relies on a built-in mapping relationship. The real-time change trend received by the real-time feedback controller is a multi-dimensional vector, and the built-in algorithm of the real-time feedback controller maps this trend vector to specific parameter adjustment amounts. The initial impact energy, standard pulse sequence, and reference action angle are superimposed with the corresponding parameter fine-tuning amounts to obtain the fine-tuned impact parameters. Superposition is an algebraic operation. For the initial impact energy, the energy compensation value is added to obtain the fine-tuned impact energy; for the pulse interval time in the standard pulse sequence, the pulse interval adjustment amount is added to obtain the new interval time; for the reference action angle, the angle offset amount is added to obtain the fine-tuned action angle. Based on the fine-tuned impact parameters, a machine-executable customized extracorporeal shock therapy instruction is generated. In some embodiments, the customized extracorporeal shock therapy instruction adopts a protocol format, which specifies in detail the waveform, timing, and spatial orientation of the energy output. The waveform is defined by the fine-tuned impact energy and pulse sequence, the timing is defined by the time parameters of the pulse sequence, and the spatial orientation is defined by the fine-tuned action angle. The process of the real-time feedback controller calculating the parameter fine-tuning amount can be represented by a feedback formula:

[0033] in: This represents a vector of parameter fine-tuning values, consisting of energy compensation, pulse interval adjustment, and angle offset. It is a preset gain matrix. A trend vector representing the real-time monitored force distribution and coordination characteristics. The reference vector represents the expected trend of feature change, and the integral term reflects the continuous cumulative response to trend deviation.

[0034] In a specific example scenario and data comparison, the basic radial extracorporeal shock wave (ECW) parameters obtained from the analysis are: initial shock energy of 5.0 joules, a standard pulse sequence of 10 pulses with a width of 2 milliseconds and an interval of 50 milliseconds, and a reference action angle of 30 degrees azimuth and 15 degrees elevation. The real-time feedback controller receives the real-time trend of force distribution and coordination characteristics. Assuming the real-time trend shows that the first component of the force distribution vector representing hyoid bone elevation force increased by 15% in three consecutive swallows, while the temporal index deviation representing the coordination of hyoid bone and pharyngeal contraction decreased, the real-time feedback controller calculates the current parameter fine-tuning based on the built-in algorithm and the above formula: energy compensation of -0.3 joules, pulse interval adjustment of -5 milliseconds, and angle offset of +2 degrees azimuth and 0 degrees elevation. Superimposing the initial parameters with the fine-tuning, the fine-tuned shock parameters are: shock energy of 4.7 joules, a pulse sequence of 10 pulses with a width of 2 milliseconds and an interval of 45 milliseconds, and an action angle of 32 degrees azimuth and 15 degrees elevation. Based on these fine-tuned impact parameters, the system generates a final customized extracorporeal shock therapy instruction. This instruction contains all the specific numerical parameters mentioned above and is encoded as drive code that can be directly parsed by the radial extracorporeal shock generator. This fine-tuning process allows the treatment plan to respond instantly to subtle improvements in the trainee's swallowing mechanics, dynamically reducing stimulation intensity and fine-tuning the stimulation rhythm and focus.

[0035] See Figure 3 This is a trend chart showing the changes in the distribution of swallowing force, used to monitor the strength changes of different muscle groups during swallowing training. It is a key visualization tool for assessing the progress of swallowing function rehabilitation. The hyoid elevation force increased from 2.1 in the first repetition to 3.2 in the tenth repetition, showing the most significant increase among the three muscle groups, reflecting continuous improvement in hyoid elevation function. The pharyngeal contraction force increased from 1.8 in the first repetition to 2.4 in the tenth repetition, with a stable overall trend, indicating a gradual increase in the strength of the pharyngeal contraction muscles. The larynx forward movement force increased from 1.5 in the first repetition to 2.0 in the tenth repetition, with a moderate increase, reflecting a gradual recovery of larynx forward movement function. All three force groups showed an overall upward trend, indicating that swallowing training has a positive effect on improving the strength of multiple muscle groups, with the hyoid elevation muscle group showing the most sensitive response and the greatest improvement. From the first to the tenth swallow, the difference in strength among the three groups gradually narrowed, suggesting that the coordination and strength balance of the swallowing muscle groups are improving.

[0036] In one embodiment of the present invention, a customized extracorporeal shock therapy command is sent to a radial extracorporeal shock generator via a data interface. The radial extracorporeal shock generator, based on the customized command, drives its energy output array to apply focused mechanical shock to the stimulation target area. The central controller of the radial extracorporeal shock generator parses the customized command, decomposing it into independent drive signals for each transducer unit in the energy output array. Each drive signal contains the ultrasonic frequency, phase, amplitude, and timing information to be emitted by each transducer unit. Each transducer unit generates an ultrasonic pulse with a specific frequency and phase according to its corresponding independent drive signal. The waveform and energy of the ultrasonic pulse are precisely controlled by the independent drive signal. By controlling the phase difference of the ultrasonic pulses emitted by each transducer unit, constructive interference is formed at the spatial location of the stimulation target area, thereby generating a high-energy-density mechanical shock focus. The mechanical shock focus applies periodic or modulated mechanical loading to the interior or surface of the stimulation target area according to the pulse sequence specified in the command. The pulse sequence determines the timing, duration, and repetition frequency of the mechanical shock focus. Understandably, this process enables the transformation of digital treatment instructions into precise physical stimuli that act on biological tissues.

[0037] In practice, while applying the impact, the sensor array simultaneously collects the trainee's real-time physiological response signals. Based on these signals, the functional deviation index is dynamically updated, and the next round of adjustments to the customized extracorporeal shock therapy command is initiated. In an example scenario, a customized extracorporeal shock therapy command is sent to the radial extracorporeal shock generator. This command contains the following key parameters: impact energy of 4.5 joules, a pulse sequence of 5 pulses, each pulse width of 1.5 milliseconds, a pulse interval of 100 milliseconds, and target coordinates of the focal point at (15, -5, 30) millimeters (with the center of the treatment head as the origin). After parsing the command, the central controller of the radial extracorporeal shock generator generates independent drive signals for an energy output array consisting of 256 transducer units. Table 1 shows examples of the independent drive signal parameters generated for four transducer units at different spatial locations after parsing.

[0038] Table 1: Example Table of Independent Drive Signal Parameters for Transducer Units

[0039] Units A, B, and C emit simultaneously to form the first impact pulse, with their phase difference calculated to achieve constructive interference at coordinates (15, -5, 30) mm. Unit D's emission timing is delayed by 100 milliseconds to generate the second pulse required by the command. It can be understood that the parameters of all 256 units are similarly calculated to ensure a focused mechanical impact focal point is formed at the target location. The mechanical impact focal point is generated periodically at the target coordinates, lasting 1.5 milliseconds each time, with an interval of 100 milliseconds. The formula for calculating the sound pressure at the focal point is:

[0040] in: This represents the sound pressure amplitude at a spatial point (x, y, z). This is the total number of transducer units. It is the first The emission amplitude of each transducer unit It is the wave number. It is the first The distance from each transducer unit to the target point (x,y,z). It is the first The driving phase of each transducer unit It is the imaginary unit. The formula describes how to control the amplitude of each unit. and phase This is achieved by superimposing sound fields at a specific point in space. During the pulse intervals between the applied focused impacts, the sensor array is triggered to perform high-speed sampling, acquiring induced electromyographic signals and tissue vibration signals generated by the impact intervention. These real-time physiological signals are sent to the processing module to dynamically calculate new functional deviation indices. For example, the peak velocity of hyoid bone elevation during the first swallow after the impact may increase from 15 mm / s before the impact to 18 mm / s; this change will be quantified as a reduction in the functional deviation index. Based on the updated and reduced functional deviation index, the system then initiates the adjustment process for the parameters of the next round of impact therapy.

[0041] See Figure 4This is a normalized time-series graph of physiological signals from the swallowing muscles, used to analyze the dynamic activation and synergistic patterns of three muscle groups—hyoid bone elevation, larynx forward movement, and pharyngeal contraction—during swallowing training. It serves as core monitoring data for swallowing function assessment and intervention program regulation. Before intervention, pharyngeal contraction was dominant; after intervention, the three muscle groups exhibited a synergistic pattern of "synchronous perturbation-synchronous recovery," indicating that swallowing movements depend on the temporal coordination of multiple muscle groups. The abrupt change at 30 seconds represents the direct physiological response to external impact, which can be used to quantitatively assess the immediate impact of the intervention on the swallowing muscles, providing feedback for optimizing parameters such as impact energy and pulse sequence. The significant rebound in signals in the later stages indicates that the swallowing muscles possess good recovery and adaptation capabilities, suggesting that the current intervention program is within a controllable range and can be further optimized to improve rehabilitation efficiency. The vertical axis represents the normalized signal, eliminating differences in the original signal amplitudes of different muscle groups, facilitating direct comparison of the relative activation levels and trends of each muscle group.

[0042] In one embodiment of the invention, during the interval of the output impact pulse from the radial extracorporeal impact generator or at a specific synchronization moment, a sensor array is triggered to perform high-speed sampling. The sampling action is precisely triggered by the synchronization signal of the radial extracorporeal impact generator. Evoked electromyographic signals, tissue vibration signals, and residual physiological signals of voluntary swallowing movements generated by the extracorporeal impact intervention are collected. The evoked electromyographic signals reflect the immediate electrical response of neuromuscular response to the impact, and the tissue vibration signals reflect the mechanical transmission of the shock wave within the tissue. The collected signals are marked as real-time response physiological signals and timestamped with the currently executed impact treatment command. Timestamping ensures that each collected signal segment can be associated with the specific impact pulse parameters corresponding to its generation. Pattern analysis is performed on the real-time response physiological signals to extract a new set of biomechanical parameters after the impact intervention. The pattern analysis method is consistent with the method used to analyze the original swallowing movement signals. The instantaneous change vector between the new biomechanical parameters after the impact intervention and the original biomechanical parameters before the impact intervention is calculated. The instantaneous change vector is obtained by subtracting the original biomechanical parameter vector from the new biomechanical parameter vector. The instantaneous change vector is projected onto the direction of the ideal correction target pattern vector. The length and direction of the projection are calculated. In some embodiments, the projection length is obtained by vector dot product and modulus operation, and the direction is indicated by the sign of the dot product result. Based on the length and direction of the projection, the value of the functional deviation index is adjusted. If the projection direction is consistent with the desired correction direction and the length is significant, the value of the functional deviation index is reduced; otherwise, corresponding adjustments are made to obtain the updated functional deviation index. It can be understood that the projection length represents the instantaneous intensity of the treatment response, and the projection direction represents whether the response develops in the desired correction direction.

[0043] In practical implementation, a specific example scenario illustrates the entire process from signal acquisition to indicator updating. During the current treatment cycle, the extracorporeal radial shock generator outputs a shock sequence containing three pulses. During the 5-millisecond interval following the first pulse, the sensor array is triggered for high-speed sampling, with a sampling duration of 20 milliseconds. The acquired real-time response physiological signals are analyzed to extract a new set of biomechanical parameters after the shock intervention. For example, the parameter representing the hyoid bone elevation velocity changes from 15.0 mm / s before the shock to 16.8 mm / s, and the parameter representing the pharyngeal contraction force changes from 1.2 units to 1.35 units. The original biomechanical parameters before the shock intervention were recorded before the shock treatment began, assumed to be [15.0, 1.2] (vector representation). The instantaneous change vector is calculated to be [+1.8, +0.15]. The ideal correction target pattern vector is set to [+20.0, +1.0] (direction parameter increases). The instantaneous change vector is projected onto the direction of the ideal correction target pattern vector, and the length of the projection is calculated. Projection components... The calculation formula is:

[0044] in: Indicates the projected length (scalar). Represents an instantaneous change vector. Represents the ideal correction target pattern vector. This represents the vector dot product operation. This represents the magnitude of the ideal correction target pattern vector. Substituting the example data... , The projected length can be calculated. It is a positive value. Since the projection direction (indicated by the positive dot product result) is consistent with the desired correction direction, and the projection length... If the impact exceeds a preset significance threshold, the system determines that the current impact has produced a positive immediate effect. The original functional deviation index is assumed to be 0.60, based on the projection length. The function deviation index was corrected and updated to 0.52.

[0045] In practice, the updated functional deviation index is input into the adaptive retrieval process, but the search sphere center is updated to the node corresponding to the currently executed impact scheme in the multidimensional impact scheme space. The adaptive retrieval is re-executed to obtain a new basic radial extracorporeal shock therapy scheme or to confirm and maintain the original scheme. If a new basic radial extracorporeal shock therapy scheme is obtained, a new customized extracorporeal shock therapy instruction is generated through a parameterized fine-tuning process, taking into account the latest real-time change trend. At the next treatment cycle or a preset adjustment time point, the new customized extracorporeal shock therapy instruction is sent to the radial extracorporeal shock therapy generator to replace or modify the currently executed instruction. In the example scenario, the updated functional deviation index of 0.52 is input into the adaptive retrieval process. The adaptive retrieval process uses the node corresponding to the currently executed impact scheme (e.g., coordinates (5,2,7)) as the new sphere center and redefines the candidate spheres based on the search radius calculated based on the new index. Optionally, since the functional deviation index decreases from 0.60 to 0.52, the calculated search radius will increase, which may lead to the retrieval of nodes with different historical correction success rates within the candidate spheres. Suppose a new node is detected, corresponding to a basic treatment plan with slightly lower impact energy and slightly higher pulse frequency. The system combines the real-time trend analysis from the latest acquired real-time response physiological signals to perform parameter fine-tuning on this new basic plan, ultimately generating a new customized extracorporeal shock therapy command. In the next treatment cycle, this new command is sent to the radial extracorporeal shock generator, replacing the original command parameters, thus achieving closed-loop, adaptive adjustment of the treatment plan.

[0046] See Figure 5 This is a time-series diagram of the physiological response to swallowing under radial extracorporeal impact intervention. It is used to quantitatively analyze the synchronous response patterns of the impact pulse and the induced electromyographic (EMG) and tissue vibration signals, providing core data for real-time feedback control in swallowing rehabilitation. Both the induced EMG and tissue vibration signals respond rapidly within milliseconds after the impact pulse, indicating that the swallowing muscles and soft tissues are highly sensitive to the mechanical stimulation of extracorporeal impact, providing a time window for real-time feedback control. The peak value of the EMG signal is significantly higher than that of the vibration signal, reflecting that neuromuscular activation is the primary response pathway of the impact intervention, while tissue vibration is a secondary manifestation of mechanical energy transfer. After the first impact (200 ms), the peak value of the EMG signal is approximately 0.8, and the peak value of the vibration signal is approximately 0.76. After the second impact (400 ms), the peak value of the EMG signal rises to 1.2, and the peak value of the vibration signal rises to 0.93, reflecting initial adaptation and enhanced activation. After the third impact (600 ms), the peak value of the EMG signal remains at 1.2, while the peak value of the vibration signal slightly decreases, suggesting that the muscle groups have entered a stable activation state, and tissue vibration is beginning to show signs of fatigue.

[0047] The above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments that can be applied to other fields. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for regulating a radial extracorporeal shock protocol in response to swallowing training data, characterized in that, The method includes: A sensor array is arranged in a pre-defined stimulation target area on the trainee's body surface. The sensor array captures multi-channel physiological signals generated when the trainee performs a standard swallowing action. Pattern analysis was performed on the multi-channel physiological signals to extract a series of biomechanical parameters characterizing the swallowing mechanics. These biomechanical parameters include the temporal characteristics, force distribution characteristics, and coordination characteristics of the swallowing action. The biomechanical parameters are input into a swallowing function assessment atlas for comparison. The swallowing function assessment atlas stores standard swallowing patterns from normal to various functional impairment states. Based on the comparison results, a quantitative functional deviation index is generated, which is used to quantify the difference between the current swallowing action and the target pattern; Based on the functional deviation index, an adaptive search is performed in a multi-dimensional impact scheme space to select a basic radial extracorporeal impact scheme. The multi-dimensional impact scheme space consists of three dimensions: impact energy, pulse sequence, and action angle. The selected basic extracorporeal shock therapy scheme is parametrically fine-tuned based on the real-time changing trends of the force distribution and coordination characteristics in the biomechanical parameters, thereby generating a customized extracorporeal shock therapy instruction.

2. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 1, characterized in that, Pattern analysis was performed on the multi-channel physiological signals to extract a series of biomechanical parameters characterizing swallowing mechanics, including: The raw physiological signals from the sensor array are preprocessed to remove power frequency interference and motion artifacts; Blind source separation was performed on the preprocessed signal to separate independent signal components directly related to hyoid bone elevation, larynx forward movement, and pharyngeal contraction. Envelope extraction and peak detection were performed on each independent signal component to determine the start time, peak time and duration of activation of each swallowing-related muscle group; Calculate the time difference between the peak times of different independent signal components as a temporal indicator of muscle coordination; By integrating the envelope area and peak amplitude of each independent signal component, a multidimensional force distribution vector is generated, which is the force distribution feature.

3. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 2, characterized in that, The biomechanical parameters are input into a swallowing function assessment atlas for comparison, including: A current swallowing feature vector is constructed from the biomechanical parameters, the current swallowing feature vector including the temporal features, force distribution features and coordination features; The current swallowing feature vector is matched with a predefined set of reference feature vectors in the swallowing function assessment atlas library; Each vector in the reference feature vector set is associated with a specific swallowing dysfunction pattern and is labeled with an ideal correction target pattern; Select the reference feature vector that has the highest similarity to the current swallowing feature vector, and determine the associated functional impairment pattern as the current primary functional impairment pattern; Calculate the Euclidean distance between the current swallowing feature vector and the ideal correction target pattern vector corresponding to the main functional impairment pattern, and use the normalized Euclidean distance as the functional deviation index.

4. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 3, characterized in that, The construction process of the swallowing function assessment atlas library includes: High-precision physiological signals and synchronous video fluorescence imaging data were collected from a large number of subjects under different swallowing tasks; Based on the video fluorescence angiography data, experts will perform functional scoring and pattern labeling for each swallowing action to establish a real functional status label; Standardized biomechanical parameters are extracted from the high-precision physiological signals to form sample feature vectors that correspond one-to-one with the functional state labels; Cluster analysis was used to classify the feature vectors of the samples, forming different swallowing action pattern clusters. The central feature vector of each pattern cluster is used as the reference feature vector, and its corresponding typical functional state is used as the label of the vector. Both are stored in the swallowing function assessment atlas library.

5. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 4, characterized in that, Based on the aforementioned functional deviation index, an adaptive search is performed in a multidimensional impact scheme space to select a basic extracorporeal radiation impaction scheme, including: A three-dimensional orthogonal coordinate system is defined as the space of the multi-dimensional impact scheme, with the three coordinate axes representing the impact energy level, pulse sequence mode encoding, and impact angle encoding, respectively. In the multidimensional impact scheme space, a series of discrete nodes are pre-calibrated, and each node corresponds to a verified and effective basic radial extracorporeal impact scheme. The functional deviation index is mapped to a search radius in the multidimensional impact scheme space; Using the historical mapping point corresponding to the current swallowing action mode in the multidimensional impact scheme space as the center of the sphere and the search radius as the radius, a candidate scheme sphere is defined. From all discrete nodes located within the candidate scheme sphere, select the node with the highest historical correction success rate of the main functional impairment pattern, and use the impact scheme corresponding to that node as the selected basic radial extracorporeal impact scheme.

6. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 5, characterized in that, The selected basic extracorporeal shock wave therapy protocol is parametrically fine-tuned to generate a customized extracorporeal shock wave therapy instruction, including: The basic radial extracorporeal impact scheme was analyzed to obtain its initial impact energy, standard pulse sequence, and reference action angle. A real-time feedback controller is established, the input of which is the real-time changing trend of the force distribution characteristics and coordination characteristics; The real-time feedback controller calculates and generates a set of parameter fine-tuning values ​​based on the changing trend of the input. The parameter fine-tuning values ​​include energy compensation value, pulse interval adjustment amount and angle offset amount. The initial impact energy, standard pulse sequence, and reference action angle are superimposed with the corresponding parameter fine-tuning amounts to obtain the fine-tuned impact parameters; Based on the fine-tuned impact parameters, a customized extracorporeal shock therapy instruction that can be executed by the machine is generated, which specifies in detail the waveform, timing and spatial orientation of the energy output.

7. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 6, characterized in that, Also includes: The customized extracorporeal shock therapy command is sent to a radial extracorporeal shock generator; The radial extracorporeal shock generator, according to the customized extracorporeal shock therapy command, drives its energy output array to apply focused mechanical shock to the stimulation target area. While the impact is being applied, the sensor array simultaneously collects the trainee's real-time response physiological signals; Based on the real-time response physiological signals, the functional deviation index is dynamically updated, and based on the updated functional deviation index, the next round of adjustment to the customized extracorporeal shock therapy command is initiated. The radial extracorporeal shock generator, according to the customized extracorporeal shock therapy command, drives its energy output array to apply focused mechanical shock to the stimulation target area, including: The central controller of the radial extracorporeal shock generator analyzes the customized extracorporeal shock therapy command and decomposes it into independent drive signals for each transducer unit in the energy output array. Each transducer unit generates an ultrasonic pulse of a specific frequency and phase according to its corresponding independent drive signal; By controlling the phase difference of the ultrasonic pulses emitted by each transducer unit, constructive interference is formed at the spatial position of the stimulation target area, thereby generating a high-energy-density mechanical impact focus. The mechanical impact focus applies periodic or modulated mechanical loading to the inside or surface of the stimulation target area according to the pulse sequence specified in the instruction.

8. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 7, characterized in that, While the impact is being applied, the sensor array simultaneously acquires the trainee's real-time physiological response signals, including: During the interval of the output impact pulse from the radial external impact generator or at a specific synchronization moment, the sensor array is triggered to perform high-speed sampling; Collect induced electromyographic signals, tissue vibration signals, and residual physiological signals of voluntary swallowing movements generated by extracorporeal impact intervention; The acquired signals are labeled as the real-time response physiological signals and timestamped with the currently executing shock therapy command. Based on the real-time response physiological signals, the functional deviation index is dynamically updated, including: Pattern analysis was performed on the real-time response physiological signals to extract a new set of biomechanical parameters after the impact intervention; Calculate the instantaneous change vector between the new biomechanical parameters after the impact intervention and the original biomechanical parameters before the impact intervention. Project the instantaneous change vector onto the direction of the ideal correction target pattern vector, and calculate the length and direction of the projection; Based on the length and direction of the projection, the value of the functional deviation index is corrected. If the projection direction is consistent with the desired correction direction and the length is significant, the value of the functional deviation index is reduced; otherwise, corresponding adjustments are made to obtain the updated functional deviation index.

9. The method for regulating a radial extracorporeal shock therapy protocol in response to swallowing training data as described in claim 8, characterized in that, Based on the updated functional deviation index, the next round of adjustments to the customized extracorporeal shock therapy instructions will be initiated, including: The updated functional deviation index is input into the adaptive retrieval process, but the search center is updated to the node corresponding to the currently executed impact scheme in the multidimensional impact scheme space. Re-execute the adaptive search to obtain a new basic extracorporeal radiological shock protocol or confirm that the original protocol is maintained. If a new basic extracorporeal shock therapy protocol is obtained, a new customized extracorporeal shock therapy instruction will be generated by combining the latest real-time trends and through a parameterized fine-tuning process. In the next treatment cycle or at a preset adjustment time, a new customized extracorporeal shock therapy command is sent to the radial extracorporeal shock generator to replace or modify the command being executed.

10. A system for regulating a radial extracorporeal shock protocol in response to swallowing training data, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the radial extracorporeal shock protocol control method for responding to swallowing training data as described in any one of claims 1 to 9.

Images