Medical equipment control method, medical equipment control device and medical equipment

Abstract

The invention provides a medical equipment control method, a medical equipment control device and medical equipment. The method is applied to the technical field of medical treatment. A radar sensor and a piezoelectric sensor are configured in a rocker of the medical equipment; the method comprises the steps that in response to an operation instruction for a rocker, a micro-motion characteristic corresponding to a digital baseband signal and a pressure characteristic corresponding to a voltage signal are obtained, the digital baseband signal is used for indicating a signal collected through a radar sensor, and the voltage signal is used for indicating a signal collected through a piezoelectric sensor; based on the micro-motion feature and the pressure feature, determining a first confidence coefficient and a second confidence coefficient, the first confidence coefficient being used for indicating that an object triggering the operation instruction is a confidence coefficient of a target object, and the second confidence coefficient being used for indicating that the operation instruction is an effective operation confidence coefficient; and determining whether to execute the operation instruction based on the first confidence coefficient and the second confidence coefficient. According to the method, the risk of mistakenly touching the operation of the medical equipment can be reduced, and the accurate control of the medical equipment is ensured.

Description

Technical Field

[0001] This application relates to the field of medical technology, and more specifically, to a control method for a medical device, a control device for a medical device, and a medical device. Background Technology

[0002] With the rapid development of the medical equipment industry, the precise control of medical equipment has become a core element affecting the safety and effectiveness of clinical diagnosis and treatment. Taking digital subtraction angiography (DSA) equipment as an example, DSA equipment, as a key device affecting surgery and the diagnosis of vascular diseases, requires medical staff to control the movement of moving parts such as the C-arm, treatment table, X-ray tube, and detectors by operating the joystick of the DSA equipment.

[0003] In related technologies, a dual-confirmation mechanism is typically used to reduce the probability of misoperation in order to achieve precise control of medical equipment. However, while the dual-confirmation mechanism can effectively avoid misoperation, it requires medical staff to perform an additional confirmation operation, reducing their operational efficiency. Therefore, how to reduce the risk of accidental operation of medical equipment and ensure precise control of it has become an urgent problem to be solved. Summary of the Invention

[0004] This application provides a control method, a control device, and the medical device itself. The method involves installing a radar sensor and a piezoelectric sensor on the joystick of the medical device. The digital baseband signal acquired by the radar sensor has strong penetration and is unaffected by light and temperature, exhibiting extremely high reliability. Based on this, by combining and analyzing the digital baseband signal with the voltage signal, it is possible to effectively distinguish between normal actions and accidental touches by medical personnel, avoiding unexpected actions during examinations, reducing the risk of accidental touches during medical device operation, and ensuring precise control of the medical device.

[0005] Firstly, a method for controlling a medical device is provided, wherein a joystick of the medical device is equipped with a radar sensor and a piezoelectric sensor; the method includes: In response to an operation command on the joystick, the system acquires a micro-motion characteristic corresponding to a digital baseband signal and a pressure characteristic corresponding to a voltage signal. The digital baseband signal indicates the signal acquired by the radar sensor, and the voltage signal indicates the signal acquired by the piezoelectric sensor. Based on the micro-motion characteristic and the pressure characteristic, a first confidence level and a second confidence level are determined. The first confidence level indicates the confidence that the object triggering the operation command is the target object, and the second confidence level indicates the confidence that the operation command is a valid operation. Based on the first confidence level and the second confidence level, the system determines whether to execute the operation command.

[0006] The above technical solution, in response to the joystick operation command, determines a first confidence level and a second confidence level based on the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal. Based on the first confidence level and the second confidence level, it determines whether to execute the operation command. Since the first confidence level indicates the confidence level that the object triggering the operation command is the target object, and the second confidence level indicates the confidence level that the operation command is a valid operation, the first confidence level and the second confidence level can be used to analyze whether the operation command is the true intention of the target object. When the operation command is the true intention of the target object, the operation command is executed, which can reduce the risk of accidental touches in the operation of medical equipment, ensure accurate control of medical equipment, and thus improve the user experience.

[0007] In conjunction with the first aspect, in some possible implementations, determining the first confidence level and the second confidence level based on the micro-motion feature and the pressure feature includes: Based on the micro-motion feature, a third confidence level is determined for the micro-motion feature as a human body feature; based on the pressure feature, a fourth confidence level is determined for the pressure feature as a target pressure feature, where the target pressure feature is the pressure feature corresponding to the target object; the third confidence level and the fourth confidence level are weighted to obtain the first confidence level.

[0008] The above technical solution determines the third confidence level of the micro-motion feature as a human body feature and the fourth confidence level of the pressure feature as a target pressure feature, and performs weighted processing on the third confidence level and the fourth confidence level to obtain the first confidence level; by comprehensively evaluating the third confidence level of the micro-motion feature of the operating joystick as a human body feature and the target feature pressure of the operating joystick as the target object, the accuracy of the first confidence level of obtaining the object that triggers the operation command is the target object is ensured.

[0009] In combination with the first aspect and the above implementation methods, in some possible implementation methods, determining the third confidence level of the micro-motion feature as a human feature based on the micro-motion feature includes: Based on the target micro-motion features in the target frequency band of the micro-motion features, the heart rate signal-to-noise ratio and multiple motion features are determined; the heart rate signal-to-noise ratio is normalized to obtain a heart rate signal-to-noise ratio confidence level; based on the multiple motion features, a continuity confidence level is determined, which is used to indicate the continuity confidence level of the trajectory that triggers the operation command; the heart rate signal-to-noise ratio confidence level and the continuity confidence level are weighted to obtain the third confidence level.

[0010] The above technical solution determines the heart rate signal-to-noise ratio (SNR) and multiple motion features based on the target frequency band of the micro-motion features. The SNR is then normalized to obtain a confidence level. Based on the multiple motion features, a continuity confidence level is determined. The SNR and continuity confidence levels are weighted to obtain a third confidence level. Finally, by using the SNR confidence level of the object executing the operation command and the continuity confidence level of the trajectory triggering the operation command, a third confidence level is obtained that the micro-motion features are human features, ensuring the accuracy of the third confidence level.

[0011] Combining the first aspect and the above implementation methods, in some possible implementation methods, the multiple motion features include the rate of change of acceleration, the consistency of motion direction, and the smoothness of motion trajectory; determining the continuity confidence level based on the multiple motion features includes: Obtain the priority weights of the multiple motion features; based on the acceleration change rate, the consistency of the motion direction, the smoothness of the motion trajectory, and the priority weights, obtain the continuity confidence level.

[0012] The above technical solution obtains the priority weights of multiple motion features, and obtains the continuity confidence based on the acceleration change rate, the consistency of motion direction, the smoothness of motion trajectory and priority weights; by weighting multiple motion features with their priority weights, the trajectory continuity of the joystick operation is determined, thereby ensuring the accuracy of the continuity confidence.

[0013] In combination with the first aspect and the above implementation methods, in some possible implementation methods, determining the fourth confidence level of the pressure feature as the target pressure feature based on the pressure feature includes: Based on the pressure characteristics, the degree of disorder in the pressure distribution on the joystick is determined; based on the pressure characteristics, a feature matching degree is obtained by matching them with preset pressure characteristics, wherein the preset pressure characteristics are used to indicate the pressure characteristics corresponding to the target object; based on the degree of disorder and the feature matching degree, a fourth confidence level is determined, wherein the degree of disorder is negatively correlated with the fourth confidence level, and the feature matching degree is positively correlated with the fourth confidence level.

[0014] The above technical solution determines the degree of disorder in the pressure distribution on the joystick based on pressure characteristics, matches the pressure characteristics with preset pressure characteristics to obtain the feature matching degree, and determines the fourth confidence level based on the degree of disorder and the feature matching degree. Since the degree of disorder is negatively correlated with the fourth confidence level and the feature matching degree is positively correlated with the fourth confidence level, determining the fourth confidence level by matching the degree of disorder in the pressure distribution with the feature matching degree corresponding to the pressure characteristics can accurately determine the object that triggers the operation command as the target object, thereby improving the precise control of medical equipment.

[0015] In combination with the first aspect and the above implementation methods, in some possible implementation methods, determining the first confidence level and the second confidence level based on the micro-motion feature and the pressure feature includes: Point cloud clustering is performed on the micro-motion features to obtain the acceleration features of hand joint movement; pressure gradient features are determined based on the pressure features; and the second confidence level is determined based on the acceleration features and the pressure gradient features.

[0016] The above technical solution performs point cloud clustering on the micro-motion features to obtain the acceleration features of the hand joint movement. Based on the pressure features, it determines the pressure gradient features, and then determines the second confidence level based on the acceleration features and pressure gradient features. By combining the micro-motion features and pressure features, it determines the acceleration features and pressure gradient features of the hand joint movement, which can determine the dynamic features of the hand that triggers the operation command, and thus determine the second confidence level, ensuring the accuracy of the second confidence level.

[0017] In combination with the first aspect and the above implementation methods, in some possible implementation methods, determining the second confidence level based on the acceleration feature and the pressure gradient feature includes: Based on the acceleration feature and the pressure gradient feature, a spatiotemporal covariance matrix is ​​constructed, which is used to indicate the degree of synchronization between the acceleration feature and the pressure gradient feature; based on the spatiotemporal covariance matrix, the maximum eigenvalue is determined; the maximum eigenvalue is numerically mapped to obtain the second confidence level.

[0018] The above technical solution constructs a spatiotemporal covariance matrix based on acceleration and pressure gradient characteristics. Based on this matrix, the largest eigenvalue is determined, and then numerically mapped to obtain a second confidence level. By constructing the spatiotemporal covariance matrix of acceleration and pressure gradient characteristics, the degree of synchronization between these characteristics is determined using the largest eigenvalue. Since the degree of synchronization is directly related to effective operation, determining the second confidence level as an effective operation based on the degree of synchronization ensures the accuracy of the second confidence level, thereby improving the precise control of medical equipment.

[0019] In combination with the first aspect and the above implementation methods, in some possible implementation methods, the piezoelectric sensor includes multiple sensing units, and the voltage signal includes multiple voltage signals collected by the multiple sensing units; The step of acquiring the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal in response to the operation command of the joystick includes: In response to the operation command of the joystick, it is determined whether there is an abnormal voltage signal among the plurality of voltage signals; when there is no abnormal voltage signal among the plurality of voltage signals, feature extraction is performed on the digital baseband signal and the plurality of voltage signals respectively to obtain the micro-motion feature and the pressure feature.

[0020] The above technical solution, in response to the operation command of the joystick, determines whether there is an abnormal voltage signal among multiple voltage signals. When there is no abnormal voltage signal among the multiple voltage signals, features are extracted from the digital baseband signal and the voltage signal respectively to obtain micro-motion features and pressure features. When the operation command of the joystick is detected, if there is no abnormal voltage signal among the multiple voltage signals, it is determined that the operation command of the joystick is not an abnormal contact. Then, the micro-motion features and voltage signals are comprehensively analyzed to determine whether to execute the operation command, thereby avoiding abnormal operation caused by abnormal contact of the joystick and ensuring precise control of the medical device.

[0021] Secondly, a control device for a medical device is provided, wherein a radar sensor and a piezoelectric sensor are configured in the joystick of the medical device; the device includes: An acquisition module is configured to acquire, in response to an operation command on the joystick, a micro-motion feature corresponding to a digital baseband signal and a pressure feature corresponding to a voltage signal, wherein the digital baseband signal indicates a signal acquired by the radar sensor and the voltage signal indicates a signal acquired by the piezoelectric sensor; a determination module is configured to determine a first confidence level and a second confidence level based on the micro-motion feature and the pressure feature, wherein the first confidence level indicates the confidence level that the object triggering the operation command is a target object and the second confidence level indicates the confidence level that the operation command is a valid operation; and an execution module is configured to determine whether to execute the operation command based on the first confidence level and the second confidence level.

[0022] In conjunction with the second aspect, in some possible implementations, the module is specifically used for: Based on the micro-motion feature, a third confidence level is determined for the micro-motion feature as a human body feature; based on the pressure feature, a fourth confidence level is determined for the pressure feature as a target pressure feature, where the target pressure feature is the pressure feature corresponding to the target object; the third confidence level and the fourth confidence level are weighted to obtain the first confidence level.

[0023] Combining the second aspect and the above implementation methods, in some possible implementation methods, the module is specifically used for: Based on the target micro-motion features in the target frequency band of the micro-motion features, the heart rate signal-to-noise ratio and multiple motion features are determined; the heart rate signal-to-noise ratio is normalized to obtain a heart rate signal-to-noise ratio confidence level; based on the multiple motion features, a continuity confidence level is determined, which is used to indicate the continuity confidence level of the trajectory that triggers the operation command; the heart rate signal-to-noise ratio confidence level and the continuity confidence level are weighted to obtain the third confidence level.

[0024] Combining the second aspect and the above implementation methods, in some possible implementations, the multiple motion characteristics include the rate of change of acceleration, the consistency of motion direction, and the smoothness of the motion trajectory; the determining module is specifically used for: Obtain the priority weights of the multiple motion features; based on the acceleration change rate, the consistency of the motion direction, the smoothness of the motion trajectory, and the priority weights, obtain the continuity confidence level.

[0025] Combining the second aspect and the above implementation methods, in some possible implementation methods, the module is specifically used for: Based on the pressure characteristics, the degree of disorder in the pressure distribution on the joystick is determined; based on the pressure characteristics, a feature matching degree is obtained by matching them with preset pressure characteristics, wherein the preset pressure characteristics are used to indicate the pressure characteristics corresponding to the target object; based on the degree of disorder and the feature matching degree, a fourth confidence level is determined, wherein the degree of disorder is negatively correlated with the fourth confidence level, and the feature matching degree is positively correlated with the fourth confidence level.

[0026] Combining the second aspect and the above implementation methods, in some possible implementation methods, the module is specifically used for: Point cloud clustering is performed on the micro-motion features to obtain the acceleration features of hand joint movement; pressure gradient features are determined based on the pressure features; and the second confidence level is determined based on the acceleration features and the pressure gradient features.

[0027] Combining the second aspect and the above implementation methods, in some possible implementation methods, the module is specifically used for: Based on the acceleration feature and the pressure gradient feature, a spatiotemporal covariance matrix is ​​constructed, which is used to indicate the degree of synchronization between the acceleration feature and the pressure gradient feature; based on the spatiotemporal covariance matrix, the maximum eigenvalue is determined; the maximum eigenvalue is numerically mapped to obtain the second confidence level.

[0028] In combination with the second aspect and the above implementation methods, in some possible implementations, the piezoelectric sensor includes multiple sensing units, and the voltage signal includes multiple voltage signals collected by the multiple sensing units; the acquisition module is specifically used for: In response to the operation command of the joystick, it is determined whether there is an abnormal voltage signal among the plurality of voltage signals; when there is no abnormal voltage signal among the plurality of voltage signals, feature extraction is performed on the digital baseband signal and the plurality of voltage signals respectively to obtain the micro-motion feature and the pressure feature.

[0029] Thirdly, a medical device is provided, including a memory and a processor, wherein the memory is used to store executable program code; and the processor is used to call and run the executable program code from the memory, causing the medical device to perform the control method of the medical device in the first aspect or any possible implementation thereof.

[0030] Fourthly, a computer-readable storage medium is provided that stores computer program code, which, when executed on a computer, causes the computer to perform the control method of the medical device in the first aspect or any possible implementation thereof.

[0031] Fifthly, a computer program product is provided, comprising: computer program code, which, when executed on a computer, causes the computer to perform the control method of the medical device in the first aspect or any possible implementation thereof. Attached Figure Description

[0032] Figure 1 This is a schematic flowchart illustrating a control method for a medical device provided in an embodiment of this application; Figure 2 This is a schematic flowchart of a digital baseband signal processing method provided in an embodiment of this application; Figure 3 This is a schematic flowchart of a voltage signal processing method provided in an embodiment of this application; Figure 4 This is a schematic flowchart illustrating another method for controlling a medical device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of a control device for a medical device provided in an embodiment of this application; Figure 6 This is a schematic diagram of the structure of a medical device provided in an embodiment of this application. Detailed Implementation

[0033] The technical solutions in this application will be clearly and thoroughly described below with reference to the accompanying drawings. In the description of the embodiments of this application, unless otherwise stated, " / " means "or," for example, A / B can mean A or B. "And / or" in the text is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Furthermore, in the description of the embodiments of this application, "multiple" refers to two or more than two.

[0034] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature.

[0035] With the rapid development of the medical equipment industry, the precise control of medical equipment has become a core element affecting the safety and effectiveness of clinical diagnosis and treatment. Taking DSA equipment as an example, as a key device affecting surgery and the diagnosis of vascular diseases, DSA equipment requires medical staff to control the movement of moving parts such as the C-arm, treatment table, X-ray tube, and detectors by operating the joystick of the DSA equipment (which can also be called the DSA control box or DSA operation box, etc.), thereby realizing the intervention or operation of the imaging process.

[0036] The purpose of preventing accidental touches on DSA devices is to avoid medical staff (or operators) from inadvertently moving the DSA device, which could lead to unexpected device movement, changes in patient position, or other potential safety risks, resulting in errors during the examination or treatment process or even harm to the patient.

[0037] In related technologies, to address accidental touches on DSA devices, various methods are typically employed to improve anti-accidental touch performance, such as dual operation confirmation mechanisms, optimized button layouts, and the addition of protective covers. While these measures reduce the probability of accidental touches to some extent, each has its own shortcomings. For example, while dual operation mechanisms can effectively prevent accidental operation, they require medical staff to perform an additional confirmation operation, which not only affects the feel of operation but also reduces the efficiency of medical staff. Furthermore, if the feel of dual operation mechanisms or optimized button layouts is not well adjusted, it can greatly cause resentment among medical staff. On the other hand, optimizing button layouts and adding protective covers cannot effectively solve the problem of accidental operation and may increase the usage costs for medical staff, significantly impacting the convenience of operation.

[0038] In view of this, this application provides a control method, a control device, and a medical device for medical equipment. A radar sensor and a piezoelectric sensor are installed on the joystick of the medical device. The digital baseband signal acquired by the radar sensor has strong penetration, is unaffected by light and temperature, and has extremely high reliability. Based on this, by combining and analyzing the digital baseband signal with the voltage signal, it is possible to effectively distinguish between normal actions and accidental touches by medical personnel, avoiding unexpected actions during examinations, reducing the risk of accidental touches during medical equipment operation, and ensuring precise control of the medical equipment.

[0039] The control method for the medical device described in this application is applied to a medical device in which a radar sensor and a piezoelectric sensor are configured in the joystick. The installation method of the radar sensor and piezoelectric sensor in the joystick will be described below.

[0040] For example, a radar sensor (or millimeter-wave radar sensor) is embedded in the top of the joystick of a medical device, with the radar sensor's antenna facing the hand contact area; that is, the millimeter-wave emission direction of the radar sensor is perpendicular to the hand contact area. The radar sensor's antenna uses a non-equal-pitch array (e.g., NLA), which improves the sidelobe suppression ratio by 15 dB compared to an equal-pitch array.

[0041] The piezoelectric sensor is mounted on the curved surface of the joystick's grip. The voltage sensor includes multiple sensing units (e.g., 15 sensing units, respectively detecting the pressure distribution at 14 knuckles and the palm). Furthermore, the radar sensor and the piezoelectric sensor are configured on the same printed circuit board substrate, achieving spatial integration of the circuitry. A high-frequency absorbing material is applied to the surface of the joystick's internal metal structure, and the joystick body surface is made of medical-grade silicone, which covers the piezoelectric sensor array.

[0042] When medical staff hold the joystick, their hands cover the top area of ​​the joystick. The digital baseband signal is collected by the radar sensor, and the voltage signal is collected by the piezoelectric sensor. This can minimize the distance between the radar sensor and the hand, and ensure the stability of the signal strength of the digital baseband signal and the voltage signal.

[0043] Taking a DSA device as an example, a 60 / 77GHz millimeter-wave radar sensor is embedded in the top of the joystick, with the radar sensor's antenna facing the hand contact area. This is combined with an array of 15 piezoelectric sensors (e.g., polyvinylidene fluoride piezoelectric sensors, PVDF; PVDF has the refractive index of an ionic liquid composite material) fitted onto the curved surface of the grip. The radar sensor and piezoelectric sensor are integrated and spatially aligned on the same PCB substrate. High-frequency absorbing material is added to the internal metal surface of the DSA device's joystick to suppress radio frequency reflection interference, and an external medical-grade silicone protective layer ensures aseptic operation compatibility.

[0044] For example, the radar sensor emits signals that can penetrate sterile plastic sleeves to detect fluctuations in blood flow in the hands of healthcare workers, respiratory rate, and micro-gesture trajectories (e.g., grip direction, swing amplitude), thereby accurately identifying the intention to actively operate the joystick. Simultaneously, an on-chip antenna focuses the beam within a 30-degree range (e.g., 15 degrees in the positive direction to 15 degrees in the negative direction), effectively shielding against interference from patients in bed more than 1 meter away using a distance-angle joint filtering algorithm. Meanwhile, a piezoelectric sensor array monitors the hand pressure distribution in real time; when the thumb pressure exceeds 2N, operation confirmation is achieved through voltage and digital baseband signals.

[0045] The above technical solution achieves spatiotemporal synchronous verification by placing the radar sensor and piezoelectric sensor on the same PCB substrate and spatially aligning them. Meanwhile, the absorbing material inside the rocker continuously suppresses multipath reflections and resonance interference from the metal cavity. This dual verification mechanism, achieved through spatiotemporal synchronization, can accurately identify the true intentions of medical personnel in complex surgical environments, significantly improving the safety and accuracy of medical equipment operation.

[0046] Figure 1 This is a schematic flowchart of a control method for a medical device provided in an embodiment of this application.

[0047] For example, Figure 1 The control method shown is applied to medical devices. For example, a DSA device.

[0048] For example, such as Figure 1 As shown, the control method 100 for the medical device includes steps S110-S140.

[0049] S110, in response to the operation command of the joystick, acquires the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal.

[0050] For example, the joystick of a medical device is equipped with a radar sensor and a piezoelectric sensor. The radar sensor collects digital baseband signals, and the piezoelectric sensor collects voltage signals. By combining and analyzing the digital baseband signals and voltage signals, it is determined whether to execute the joystick's operation command.

[0051] Optionally, image data can be acquired using a camera on the joystick, and the presence of an operation command for the joystick can be determined based on the image data. Alternatively, the presence of an operation command for the joystick can be determined using digital baseband signals, voltage signals, and voice data. The method for determining the operation command for the joystick can be determined according to the actual situation and is not specifically limited here.

[0052] Optionally, the operation commands for the joystick may include, but are not limited to, movement operations, turning operations, etc.

[0053] In one example, a pressure sensor configured in the joystick collects a pressure signal and determines whether the pressure signal is greater than a preset pressure signal. When the pressure signal is less than or equal to the preset pressure signal, the system continuously uses the pressure signal to determine whether there is an operation command for the joystick. When the pressure signal is greater than the preset pressure signal, it is determined that there is an operation command for the joystick.

[0054] Optionally, a preset pressure signal is used to indicate the presence of a pressure signal that touches the joystick. The preset pressure signal can be 2N, 3N, etc.

[0055] In another example, a radar sensor configured in the joystick acquires a digital baseband signal. The system determines whether the digital baseband signal is greater than a preset baseband signal. When the digital baseband signal is less than or equal to the preset baseband signal, the system continuously uses the digital baseband signal to determine if there is an operation command for the joystick. When the digital baseband signal is greater than the preset baseband signal, it is determined that there is an operation command for the joystick. Optionally, the preset baseband signal is used to indicate the presence of a baseband signal indicating contact with the joystick.

[0056] In another example, image data is captured by a camera configured in the joystick. This image data is then input into an image recognition model, which outputs whether any joystick operation is detected in the image data. If a joystick operation is detected in the image data, it is determined that an operation command for the joystick has been detected. Optionally, the image recognition model is a model trained using sample data.

[0057] Upon detecting an operation command on the joystick of a medical device, the operation command is analyzed to prevent abnormal operation of the joystick.

[0058] For example, the piezoelectric sensor includes multiple sensing units, and the voltage signal includes multiple voltage signals collected by the multiple sensing units. In response to the operation command of the joystick, it is determined whether there is an abnormal voltage signal among the multiple voltage signals; when there is no abnormal voltage signal among the multiple voltage signals, feature extraction is performed on the digital baseband signal and the voltage signal respectively to obtain micro-motion features and pressure features.

[0059] Optionally, an abnormal voltage signal indicates that a single voltage signal among multiple voltage signals is significantly larger than the others, and this single voltage signal exceeds the upper pressure threshold. For example, a radar sensor may include 15 sensing units, where a single sensing unit acquires a voltage signal that is larger than the voltage signals acquired by the other 14 sensing units, and this single sensing unit's voltage signal exceeds the upper pressure threshold. Furthermore, an abnormal voltage signal can be referred to as a single-point full-scale pressure.

[0060] For example, Shannon entropy of pressure distribution at multiple points is calculated based on multiple voltage signals. When the change in Shannon entropy exceeds a preset change, an abnormal voltage signal is identified among the multiple voltage signals. Optionally, the preset change can be 1.5, 1.6, etc., and is not specifically limited here.

[0061] Optionally, the multiple sensing units include 15 sensing units arranged according to the joint points. Of course, the multiple sensing units can also be 5, 14, 20, etc., and there is no specific limitation here.

[0062] Upon detecting an operation command on the joystick, it is determined whether a single point of full-scale pressure exists among the multiple voltage signals collected by multiple sensing units (or whether a single voltage signal collected by a single sensing unit exceeds the upper pressure threshold). If no single point of full-scale pressure exists among the multiple voltage signals collected by multiple sensing units, feature extraction is performed on the digital baseband signal collected by the radar sensor and the multiple voltage signals collected by the piezoelectric sensor to obtain the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signals. If a single point of full-scale pressure exists among the multiple voltage signals collected by multiple sensing units, it is determined that the operation on the joystick is an abnormal contact, triggering the hardware emergency stop circuit, and thus stopping the operation on the joystick.

[0063] The above technical solution, in response to the operation command of the joystick, determines whether there is an abnormal voltage signal among multiple voltage signals. When there is no abnormal voltage signal among the multiple voltage signals, features are extracted from the digital baseband signal and the voltage signal respectively to obtain micro-motion features and pressure features. When the operation command of the joystick is detected, if there is no abnormal voltage signal among the multiple voltage signals, it is determined that the operation command of the joystick is not an abnormal contact. Then, the micro-motion features and voltage signals are comprehensively analyzed to determine whether to execute the operation command, thereby avoiding abnormal operation caused by abnormal contact of the joystick and ensuring precise control of the medical device.

[0064] Furthermore, upon determining that an operation command to the joystick has been detected, the micro-motion characteristics corresponding to the digital baseband signal acquired by the radar sensor and the pressure characteristics corresponding to the voltage signal acquired by the piezoelectric sensor are obtained.

[0065] Optionally, the micro-motion characteristics corresponding to the digital baseband signal are used to indicate the minute movements of the hand in the top region of the joystick and the physiological vibration patterns. The pressure characteristics corresponding to the voltage signal are used to indicate the force applied when the hand contacts the joystick handle and the corresponding patterns. By comprehensively analyzing data from both static force and dynamic motion dimensions, it is possible to accurately distinguish between intentional operations and accidental touches by medical personnel.

[0066] In one example, based on the digital baseband signal acquired by a radar sensor, the blood flow and respiratory rate that trigger the operation command are determined, and then human verification is performed. Simultaneously, relevant data for each joint point in the human body are determined. Specifically, the digital baseband signal acquired by the radar sensor is subjected to a Doppler Fast Fourier Transform to obtain Doppler matrix data. Micro-motion features are then extracted from the Doppler matrix data. Based on this, phase unwrapping processing is performed on the micro-motion features to avoid the problem of periodic phase transitions, resulting in processed data. This processed data is then bandpass filtered using a 0.5-2Hz bandpass filter to obtain filtered data. The filtered data is then refined to obtain refined data. Based on the refined data, the heart rate signal-to-noise ratio (SNR) is determined, and the SNR is normalized to obtain the heart rate SNR confidence level. Of course, by using a preset clustering algorithm to perform point cloud clustering on the micro-motion features, clustered data is obtained. The joint point data of the previous and next frames in the clustered data is determined based on the joint point coordinates. Then, the acceleration features of the hand joint movement are obtained based on the joint point data of the previous and next frames.

[0067] In another example, a voltage signal is acquired using a piezoelectric sensor. This voltage signal is then dynamically baseline-calibrated using a pre-defined calibration algorithm to obtain a calibrated pressure signal. Features are extracted from this calibrated pressure signal to obtain pressure features. Based on this, a sliding window integration is performed on the pressure features to obtain pressure gradient features. Alternatively, a spatial difference algorithm can be used to spatially analyze the pressure features to obtain the pressure distribution entropy value. Of course, feature matching can also be performed between the pressure features and the pressure features corresponding to the target object to obtain the feature matching degree.

[0068] S120, based on micro-motion characteristics and pressure characteristics, determines the first confidence level and the second confidence level.

[0069] For example, the first confidence level indicates the confidence that the object triggering the operation instruction is the target object, ensuring that the operator of the medical device is a specific person. Optionally, the object refers to the object operating the medical device, while the target object refers to the specified object operating the medical device. For example, when the medical device is a DSA device, the object can be a nurse, patient, doctor, etc., but the target object is a doctor.

[0070] For example, the second confidence level is used to indicate the confidence level that the operation command is a valid operation, where a valid operation refers to an operation with genuine intent. By comprehensively analyzing the first and second confidence levels, the probability that the object operating the joystick of the medical device is the target object and is intentionally operated can be determined.

[0071] It should be noted that, in response to the joystick operation command of the medical device, the digital baseband signal and voltage signal are spatiotemporally aligned based on the timestamp of the signals collected by the radar sensor and piezoelectric sensor, according to the hardware pulse per second (PPS) marker, establishing a mapping relationship between the digital baseband signal and the voltage signal. By processing the baseband signal data and voltage signal with the same spatiotemporal data, a first confidence level that the object triggering the operation command is the target object, and a second confidence level that the operation command is a valid operation are obtained. Then, based on the first confidence level and the second confidence level, it is determined whether to execute the operation command.

[0072] The following section will introduce the method for determining the first confidence level of the object that triggers the operation command as the target object.

[0073] In one example, the micro-motion features and pressure features are input into the first prediction model to obtain the candidate confidence scores of the objects that trigger the operation command as candidate objects; the candidate objects are then matched with the target objects to obtain the first confidence scores of the objects that trigger the operation command as target objects. The first prediction model is a trained prediction model.

[0074] In another example, based on micro-motion features, the third confidence level of micro-motion features as human body features is determined; based on pressure features, the fourth confidence level of pressure features as target pressure features is determined, and the target pressure features are the pressure features corresponding to the target object; and the third confidence level and the fourth confidence level are weighted to obtain the first confidence level.

[0075] Optionally, human features are used to indicate that the object being operated on is a human. A third confidence level for determining the micro-motion feature as a human feature is used to indicate the confidence level that the micro-motion feature of the operating joystick is a human.

[0076] Optionally, the target pressure feature is the pressure feature corresponding to the target object.

[0077] The above technical solution determines the third confidence level of the micro-motion feature as a human body feature and the fourth confidence level of the pressure feature as a target pressure feature, and performs weighted processing on the third confidence level and the fourth confidence level to obtain the first confidence level; by comprehensively evaluating the third confidence level of the micro-motion feature of the operating joystick as a human body feature and the target feature pressure of the operating joystick as the target object, the accuracy of the first confidence level of obtaining the object that triggers the operation command is the target object is ensured.

[0078] For example, based on the target micro-motion features in the target frequency band of the micro-motion features of the joystick operation command, the heart rate signal-to-noise ratio (SNR) and multiple motion features are determined. The heart rate SNR is normalized to obtain a heart rate SNR confidence score. Based on multiple motion features, a continuity confidence score is determined. The continuity confidence score is used to indicate the continuity confidence score of the trajectory that triggers the operation command. The heart rate SNR confidence score and the continuity confidence score are then weighted to obtain a third confidence score.

[0079] Optionally, the target frequency band is the 0.5Hz to 2Hz band.

[0080] For example, in response to the operation command of the joystick, the micro-motion features corresponding to the digital baseband signal are acquired, the micro-motion features are subjected to phase unwinding processing to obtain the processed features, and the processed features are filtered by a bandpass filter (e.g., a 0.5-2Hz bandpass filter) to obtain the filtered features. Then, the filtered features are subjected to linear frequency modulated Z-transform (CZT) to obtain the heart rate signal-to-noise ratio and multiple motion features. Finally, the heart rate signal-to-noise ratio is normalized to obtain the heart rate signal-to-noise ratio confidence level.

[0081] For example, the confidence level of heart rate signal-to-noise ratio can be expressed as follows: ; ; in, This represents the heart rate signal-to-noise ratio. Indicates the effective power of the signal. Indicates the effective power of the noise. This represents the confidence level of the heart rate signal-to-noise ratio.

[0082] Optionally, motion features are used to indicate the motion attributes displayed by the hand when the operation command is triggered. Motion features may include, but are not limited to, the rate of change of acceleration, the consistency of the motion direction, and the smoothness of the motion trajectory, etc., without specific limitations here. For example, based on multiple motion features, the confidence level of the trajectory continuity of the triggered operation command is determined.

[0083] The above technical solution determines the heart rate signal-to-noise ratio (SNR) and multiple motion features based on the target frequency band of the micro-motion features. The SNR is then normalized to obtain a confidence level. Based on the multiple motion features, a continuity confidence level is determined. The SNR and continuity confidence levels are weighted to obtain a third confidence level. Finally, by using the SNR confidence level of the object executing the operation command and the continuity confidence level of the trajectory triggering the operation command, a third confidence level is obtained that the micro-motion features are human features, ensuring the accuracy of the third confidence level.

[0084] When multiple motion features include the rate of change of acceleration, consistency of motion direction, and smoothness of motion trajectory, the continuity confidence of the trajectory triggering the joystick operation command is determined based on the rate of change of acceleration, consistency of motion direction, and smoothness of motion trajectory. Specifically, the priority weights of multiple motion features are obtained, and the continuity confidence is obtained based on the rate of change of acceleration, consistency of motion direction, smoothness of motion trajectory, and priority weights.

[0085] Optionally, the priority weights of multiple motion features are as follows: a weight of 0.4 for the rate of change of acceleration (or low Jerk), a weight of 0.3 for the consistency of motion direction, and a weight of 0.3 for the smoothness of the motion trajectory. Alternatively, the priority weights of multiple motion features can be as follows: a weight of 0.5 for the rate of change of acceleration, a weight of 0.3 for the consistency of motion direction, and a weight of 0.2 for the smoothness of the motion trajectory. The priority weights of multiple motion features can be determined according to the actual situation and are not specifically limited here.

[0086] The above technical solution obtains the priority weights of multiple motion features, and obtains the continuity confidence based on the acceleration change rate, the consistency of motion direction, the smoothness of motion trajectory and priority weights; by weighting multiple motion features with their priority weights, the trajectory continuity of the joystick operation is determined, thereby ensuring the accuracy of the continuity confidence.

[0087] For example, based on the target micro-motion features in the target frequency band of the micro-motion features of the joystick operation commands, the heart rate signal-to-noise ratio (SNR) and three motion features are determined, and the heart rate SNR is normalized to obtain the heart rate SNR confidence score. Then, based on the rate of change of acceleration, the consistency of motion direction, the smoothness of the motion trajectory, and priority weights, the continuity confidence score is obtained. For example, the expression for the continuity confidence score can be represented as follows: ; in, Indicates the continuity confidence level. Indicates the rate of change of acceleration. Indicates consistency in the direction of motion. Indicates the smoothness of the motion trajectory.

[0088] Given a determined confidence level for the heart rate signal-to-noise ratio (SNR) and continuity, a weighted average is applied to the SNR and continuity confidence levels to obtain a third confidence level. For example, the expression for the third confidence level can be as follows: ; in, Indicates the third confidence level. This represents the confidence level of the heart rate signal-to-noise ratio. This indicates the confidence level for continuity.

[0089] For example, based on the pressure characteristics of the joystick operation command, the degree of disorder of the pressure distribution on the joystick is determined. The pressure characteristics are matched with preset pressure characteristics to obtain the feature matching degree. The preset pressure characteristics are used to indicate the pressure characteristics corresponding to the target object. Based on the degree of disorder and the feature matching degree, a fourth confidence level is determined. The degree of disorder is negatively correlated with the fourth confidence level, and the feature matching degree is positively correlated with the fourth confidence level.

[0090] Optionally, the degree of disorder in the pressure distribution is used to indicate the degree of disorder in the pressure distribution on the joystick caused by the action that triggers the operation command; the degree of disorder in the pressure distribution can be represented by the pressure distribution entropy value.

[0091] Optionally, the preset pressure feature is used to indicate the pressure feature corresponding to the target object, that is, to characterize the pressure distribution habit of the target object when holding the joystick; by matching the pressure feature with the preset pressure feature, the consistency between the pressure distribution and the preset pressure distribution can be determined, thereby determining the probability that the object that triggers the operation command is the target object.

[0092] This means that preset pressure features can be updated. Specifically, the preset pressure features are updated using Dynamic Time Warping (DTW). The DTW algorithm aligns the real-time pressure curve with the pre-stored template. By comparing the feature matching degree between the real-time pressure curve and the pressure features in the pre-stored template, the target object is updated, thereby updating the preset pressure features corresponding to the target object. For example, the pre-stored template includes the pressure features of object 1 and object 2. The historical target object is object 1. Due to personnel changes, the updated target object is object 2. When object 2 operates the joystick, the target object is dynamically updated, thereby updating the preset pressure features.

[0093] The above technical solution determines the degree of disorder in the pressure distribution on the joystick based on pressure characteristics, matches the pressure characteristics with preset pressure characteristics to obtain the feature matching degree, and determines the fourth confidence level based on the degree of disorder and the feature matching degree. Since the degree of disorder is negatively correlated with the fourth confidence level and the feature matching degree is positively correlated with the fourth confidence level, determining the fourth confidence level by matching the degree of disorder in the pressure distribution with the feature matching degree corresponding to the pressure characteristics can accurately determine the object that triggers the operation command as the target object, thereby improving the precise control of medical equipment.

[0094] For example, the expression for the fourth confidence level can be represented as follows: ; in, denoted by , C represents the degree of disorder in the pressure distribution on the joystick, and S represents the feature matching degree.

[0095] Given the third and fourth confidence levels, a weighted average is applied to obtain the first confidence level. The expression for the first confidence level can be represented as follows: ; in, Indicates the first confidence level. Indicates the third confidence level. This indicates the fourth confidence level.

[0096] The method for determining the second confidence level of an operation instruction as a valid operation will be introduced below.

[0097] In one example, the micro-motion features and pressure features are input into the second prediction model to obtain a second confidence level that the operation instruction is a valid operation. The second prediction model is a trained prediction model.

[0098] In another example, point cloud clustering is performed on the micro-motion features to obtain the acceleration features of hand joint movement. Based on the pressure features, the pressure gradient features are determined; and based on the acceleration features and the pressure gradient features, the second confidence level is determined.

[0099] Optionally, the acceleration characteristics of hand joint movements are used to indicate the trend of hand joint movements that trigger the operation command. These acceleration characteristics include lateral and longitudinal acceleration. Pressure gradient characteristics are used to indicate the pressure distribution of the hand that triggers the operation command. These characteristics include temporal and spatial pressure variations. The degree of synchronization between acceleration and pressure distribution can be determined through the acceleration and pressure gradient characteristics of hand joint movements.

[0100] Specifically, point cloud clustering is performed on the micro-motion features to obtain cluster features, and joint tracking is performed on the cluster features to obtain the acceleration features of hand joint movement. A sliding window integration is performed on the pressure features to obtain pressure gradient features. Then, based on the acceleration and pressure gradient features, a second confidence level is determined to confirm that the operation command is a valid operation.

[0101] The above technical solution performs point cloud clustering on the micro-motion features to obtain the acceleration features of the hand joint movement. Based on the pressure features, it determines the pressure gradient features, and then determines the second confidence level based on the acceleration features and pressure gradient features. By combining the micro-motion features and pressure features, it determines the acceleration features and pressure gradient features of the hand joint movement, which can determine the dynamic features of the hand that triggers the operation command, and thus determine the second confidence level, ensuring the accuracy of the second confidence level.

[0102] For example, a spatiotemporal covariance matrix is ​​constructed based on acceleration features and pressure gradient features. The spatiotemporal covariance matrix is ​​used to indicate the degree of synchronization between acceleration features and pressure gradient features. Based on the spatiotemporal covariance matrix, the largest eigenvalue is determined. Then, the largest eigenvalue is numerically mapped to obtain the second confidence level.

[0103] Optionally, the spatiotemporal covariance matrix is ​​used to indicate the degree of synchronization between acceleration features and pressure gradient features. The higher the degree of synchronization between acceleration features and pressure gradient features, the higher the probability that the operation command represents a valid operation intention; the lower the degree of synchronization between acceleration and pressure features, the higher the probability that the operation command is a false trigger.

[0104] The above technical solution constructs a spatiotemporal covariance matrix based on acceleration and pressure gradient characteristics. Based on this matrix, the largest eigenvalue is determined, and then numerically mapped to obtain a second confidence level. By constructing the spatiotemporal covariance matrix of acceleration and pressure gradient characteristics, the degree of synchronization between these characteristics is determined using the largest eigenvalue. Since the degree of synchronization is directly related to effective operation, determining the second confidence level as an effective operation based on the degree of synchronization ensures the accuracy of the second confidence level, thereby improving the precise control of medical equipment.

[0105] Assume the acceleration characteristics of hand joint motion are as follows: The pressure gradient characteristics are The piezoelectric sensor includes 15 sensing units, and collects 15 sets of spatially aligned acceleration and pressure gradient features based on the 15 sensing units and the radar sensor. This represents the acceleration of the joint motion in the x-direction. This represents the acceleration of the joint motion in the y-direction. This indicates how pressure changes over time; for example, a thumb-pressed sensor takes 0.5 seconds for the pressure to change from 1N to 3N. . This indicates how pressure varies with space; for example, the pressure sensor for the thumb is 3N, the pressure sensor for the index finger next to it is 2N, and the distance between the thumb and index finger is 1cm. .

[0106] Specifically, determine the acceleration of the joint motion in the x-direction. Acceleration of joint motion in the y-direction Stress changes over time Pressure changes with space The mean of the values ​​is used to construct a 4*4 spatiotemporal covariance matrix, which can determine the degree of synchronization between acceleration features and pressure gradient features.

[0107] Furthermore, the largest eigenvalue is determined based on the spatiotemporal covariance matrix. The largest eigenvalue reflects the maximum correlation strength between the acceleration feature and the pressure gradient feature. 5% of the largest eigenvalue is then determined as the second confidence level.

[0108] S130, based on the first confidence level and the second confidence level, determines whether to execute the operation instruction.

[0109] For example, based on the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal, a first confidence level is obtained that the object triggering the operation command is the target object, and a second confidence level is obtained that the operation command is a valid operation. Based on the first and second confidence levels, it is determined whether to execute the operation command. When it is determined to execute the operation command, the medical device is controlled to operate according to the operation command; when it is determined not to execute the operation command, a warning message is output to indicate that there has been an accidental touch of the joystick on the medical device.

[0110] In one example, it is determined whether a first confidence level is greater than a first threshold and whether a second confidence level is greater than a second threshold. If the first confidence level is greater than the first threshold and the second confidence level is greater than the second threshold, it is determined to execute the operation instruction; if the first confidence level is less than or equal to the first threshold, or the second confidence level is less than or equal to the second threshold, it is determined not to execute the operation instruction.

[0111] Optionally, the first threshold can be 0.7, 0.8, etc. Preferably, the first threshold is 0.7; this threshold is obtained from historical medical data statistics, and in actual operation, the first threshold of 0.7 covers 95% of effective operations.

[0112] Optionally, the second threshold can be 0.7, 0.8, etc., without specific limitations.

[0113] In another example, it is determined whether a first confidence level is greater than a first threshold or a second confidence level is greater than a second threshold. If the first confidence level is greater than the first threshold or the second confidence level is greater than the second threshold, it is determined to execute the operation instruction; if the first confidence level is less than or equal to the first threshold and the second confidence level is less than or equal to the second threshold, it is determined not to execute the operation instruction.

[0114] In another example, the first confidence level and the second confidence level are weighted to obtain the target confidence level; it is determined whether the target confidence level is greater than the third threshold. If the target confidence level is greater than the third threshold, the operation instruction is executed; if the target confidence level is less than or equal to the third threshold, the operation instruction is not executed.

[0115] The above technical solution, in response to the joystick operation command, determines a first confidence level and a second confidence level based on the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal. Based on the first confidence level and the second confidence level, it determines whether to execute the operation command. Since the first confidence level indicates the confidence level that the object triggering the operation command is the target object, and the second confidence level indicates the confidence level that the operation command is a valid operation, the first confidence level and the second confidence level can be used to analyze whether the operation command is the true intention of the target object. When the operation command is the true intention of the target object, the operation command is executed, which can reduce the risk of accidental touches in the operation of medical equipment, ensure accurate control of medical equipment, and thus improve the user experience.

[0116] Figure 2 This is a schematic flowchart of a digital baseband signal processing method provided in an embodiment of this application.

[0117] For example, such as Figure 2 As shown, the digital baseband signal processing method 200 includes steps S201-S210.

[0118] S201, acquire the digital baseband signal collected by the radar sensor.

[0119] For example, digital baseband signals are obtained by acquiring and converting data using radar sensors.

[0120] S202 performs joint estimation of the digital baseband signal to obtain Doppler matrix data.

[0121] For example, when a digital baseband signal is acquired through a radar sensor, the digital baseband signal is jointly estimated to obtain Doppler matrix data. Specifically, the digital baseband signal is subjected to a Doppler Fast Fourier Transform (Doppler FFT) to obtain the Doppler matrix data.

[0122] S203, perform micro-motion feature extraction on Doppler matrix data to obtain micro-motion features.

[0123] For example, micro-motion features are extracted from Doppler matrix data to obtain micro-motion features.

[0124] S204, perform phase unwinding processing on the micro-motion feature to obtain the processed data.

[0125] For example, performing phase unwrapping processing on the micro-motion features avoids the problem of periodic phase jumps and obtains the true phase change trend.

[0126] S205, the processed data is filtered to obtain filtered data.

[0127] For example, the processed data is bandpass filtered using a 0.5-2Hz bandpass filter to obtain filtered data.

[0128] S206 refines the filtered data to obtain refined data.

[0129] For example, the filtered data is subjected to CAT refinement to obtain refined data.

[0130] S207 determines the heart rate signal-to-noise ratio based on the refined data.

[0131] For example, based on the refined data, peak data, heart rate signal-to-noise ratio, etc. can be obtained.

[0132] S208 normalizes the heart rate signal-to-noise ratio to obtain the confidence level of the heart rate signal-to-noise ratio.

[0133] For example, the heart rate signal-to-noise ratio (SNR) is normalized and mapped to a range of 0 to 1 to obtain the heart rate SNR confidence level.

[0134] S209, perform point cloud clustering on the micro-motion features to obtain clustered data.

[0135] For example, point cloud clustering is performed on the micro-motion features using a preset clustering algorithm to obtain clustered data. Optionally, the preset clustering algorithm includes, but is not limited to, K-Means clustering, hierarchical clustering, HDBSCAN, etc., and is not specifically limited here.

[0136] S210, perform joint tracking on the clustered data to obtain the acceleration characteristics of hand joint movement.

[0137] For example, the joint point data of the preceding and following frames in the cluster data are determined based on the joint point coordinates, and then the acceleration characteristics of the hand joint movement are obtained based on the joint point data of the preceding and following frames.

[0138] The above technical solution can ensure data consistency by analyzing digital baseband signals from the same data source, and then use the consistent data to determine the accuracy of the acceleration characteristics of hand joint movement and the confidence level of heart rate signal-to-noise ratio.

[0139] Figure 3 This is a schematic flowchart of a voltage signal processing method provided in an embodiment of this application.

[0140] For example, such as Figure 3 As shown, the voltage signal processing method 300 includes steps S310-S360.

[0141] S310 acquires the voltage signal collected by the piezoelectric sensor.

[0142] For example, voltage signals are acquired using a piezoelectric sensor.

[0143] S320 performs dynamic baseline calibration on the voltage signal to obtain the calibrated pressure signal.

[0144] For example, the voltage signal is dynamically baseline-calibrated using a preset calibration algorithm to obtain the calibrated pressure signal.

[0145] S330 extracts features from the calibrated pressure signal to obtain pressure characteristics.

[0146] For example, feature extraction is performed on the calibrated pressure signal to obtain pressure features.

[0147] S340, based on the pressure features, performs sliding window integration to obtain the pressure gradient features.

[0148] For example, a sliding window integration is performed on the pressure feature to obtain the pressure gradient feature. Optionally, the length of the sliding window can be 10ms, 30ms, 100ms, etc., and is not specifically limited here.

[0149] S350, based on pressure characteristics, performs spatial analysis to obtain the pressure distribution entropy value.

[0150] For example, the pressure characteristics are spatially analyzed using a spatial difference algorithm to obtain the pressure distribution entropy value.

[0151] S360 matches the pressure features with the pressure features corresponding to the target object to obtain the feature matching degree.

[0152] For example, feature matching is performed between the pressure feature and the pressure feature corresponding to the target object to obtain the feature matching degree. Alternatively, similarity calculation is performed between the pressure feature and the pressure feature corresponding to the target object, and the similarity calculation result is determined as the feature matching degree.

[0153] The above technical solution can ensure data consistency by analyzing voltage signals from the same data source, and then determine the accuracy of pressure gradient characteristics and characteristic matching degree through consistent data.

[0154] Figure 4 This is a schematic flowchart of another control method for a medical device provided in the embodiments of this application.

[0155] For example, the joystick of a medical device is equipped with a radar sensor and a piezoelectric sensor. The radar sensor collects digital baseband signals, and the piezoelectric sensor collects voltage signals. By combining and analyzing the digital baseband signals and voltage signals, it is determined whether to execute the joystick's operation command.

[0156] For example, such as Figure 4 As shown, the control method 400 for the medical device includes steps S401-S415.

[0157] S401, in response to the operation command of the joystick, determines whether there is an abnormal voltage signal among multiple voltage signals.

[0158] Optionally, image data can be acquired using a camera on the joystick, and the presence of an operation command for the joystick can be determined based on the image data. Alternatively, the presence of an operation command for the joystick can be determined using digital baseband signals, voltage signals, and voice data. The method for determining the operation command for the joystick can be determined according to the actual situation and is not specifically limited here.

[0159] Optionally, the operation commands for the joystick may include, but are not limited to, movement operations, turning operations, etc.

[0160] Upon detecting an operation command on the joystick of a medical device, the operation command is analyzed to prevent abnormal operation of the joystick.

[0161] For example, the piezoelectric sensor includes multiple sensing units, and the voltage signal includes multiple voltage signals collected by the multiple sensing units. In response to an operation command on the joystick, it is determined whether there is an abnormal voltage signal among the multiple voltage signals.

[0162] Optionally, an abnormal voltage signal indicates that a single voltage signal among multiple voltage signals is significantly larger than the others, and this single voltage signal exceeds the upper pressure threshold. For example, a radar sensor may include 15 sensing units, where a single sensing unit acquires a voltage signal that is larger than the voltage signals acquired by the other 14 sensing units, and this single sensing unit's voltage signal exceeds the upper pressure threshold. Furthermore, an abnormal voltage signal can be referred to as a single-point full-scale pressure.

[0163] For example, Shannon entropy of pressure distribution at multiple points is calculated based on multiple voltage signals. When the change in Shannon entropy exceeds a preset change, an abnormal voltage signal is identified among the multiple voltage signals. Optionally, the preset change can be 1.5, 1.6, etc., and is not specifically limited here.

[0164] Optionally, the multiple sensing units include 15 sensing units arranged according to the joint points. Of course, the multiple sensing units can also be 5, 14, 20, etc., and there is no specific limitation here.

[0165] S402 triggers the hardware emergency stop circuit when an abnormal voltage signal is present.

[0166] For example, upon detecting an operation command on the joystick, it is determined whether there is a single point of full-scale pressure among the multiple voltage signals collected by multiple sensing units (or whether there is a single voltage signal collected by a single sensing unit that is greater than the upper pressure threshold). If a single point of full-scale pressure is found among the multiple voltage signals collected by multiple sensing units, it is determined that the operation on the joystick is an abnormal contact, triggering the hardware emergency stop circuit, and thus stopping the operation on the joystick.

[0167] S403, when there is no abnormal voltage signal, performs feature extraction on the digital baseband signal and voltage signal respectively to obtain micro-motion features and pressure features.

[0168] For example, when an operation command to the joystick is detected, it is determined whether there is a single point full-scale pressure among the multiple voltage signals collected by multiple sensing units. If there is no single point full-scale pressure among the multiple voltage signals collected by multiple sensing units, feature extraction is performed on the digital baseband signal collected by the radar sensor and the multiple voltage signals collected by the piezoelectric sensor respectively to obtain the micro-motion feature corresponding to the digital baseband signal and the pressure feature corresponding to the voltage signal.

[0169] Optionally, the micro-motion characteristics corresponding to the digital baseband signal are used to indicate the minute movements of the hand in the top region of the joystick and the physiological vibration patterns. The pressure characteristics corresponding to the voltage signal are used to indicate the force applied when the hand contacts the joystick handle and the corresponding patterns. By comprehensively analyzing data from both static force and dynamic motion dimensions, it is possible to accurately distinguish between intentional operations and accidental touches by medical personnel.

[0170] Optionally, the determination of the micro-motion characteristics and pressure characteristics can refer to the methods in the aforementioned disclosed embodiments, and will not be repeated here.

[0171] S404 performs point cloud clustering on the micro-motion features to obtain the acceleration features of hand joint movement, and determines the pressure gradient features based on the pressure features.

[0172] For example, point cloud clustering is performed on the micro-motion features to obtain clustered features, and joint tracking is performed on the clustered features to obtain the acceleration features of hand joint movement. Additionally, sliding window integration is performed on the pressure features to obtain pressure gradient features.

[0173] Optionally, the acceleration characteristics of hand joint movements are used to indicate the trend of hand joint movements that trigger the operation command. These acceleration characteristics include lateral and longitudinal acceleration. Pressure gradient characteristics are used to indicate the pressure distribution of the hand that triggers the operation command. These characteristics include temporal and spatial pressure variations. The degree of synchronization between acceleration and pressure distribution can be determined through the acceleration and pressure gradient characteristics of hand joint movements.

[0174] S405, based on acceleration and pressure gradient characteristics, constructs a spatiotemporal covariance matrix.

[0175] For example, a spatiotemporal covariance matrix is ​​constructed based on acceleration features and pressure gradient features. The spatiotemporal covariance matrix is ​​used to indicate the degree of synchronization between acceleration features and pressure gradient features.

[0176] Optionally, the spatiotemporal covariance matrix is ​​used to indicate the degree of synchronization between acceleration features and pressure gradient features. The higher the degree of synchronization between acceleration features and pressure gradient features, the higher the probability that the operation command represents a valid operation intention; the lower the degree of synchronization between acceleration and pressure features, the higher the probability that the operation command is a false trigger.

[0177] Alternatively, the spatiotemporal covariance matrix can be constructed in accordance with the aforementioned disclosed embodiments, and will not be repeated here.

[0178] S406, based on the spatiotemporal covariance matrix, determines the largest eigenvalue, performs numerical mapping on the largest eigenvalue, and obtains the second confidence level.

[0179] For example, a spatiotemporal covariance matrix is ​​constructed based on acceleration features and pressure gradient features. The spatiotemporal covariance matrix is ​​used to indicate the degree of synchronization between acceleration features and pressure gradient features. Based on the spatiotemporal covariance matrix, the largest eigenvalue is determined. Then, the largest eigenvalue is numerically mapped to obtain the second confidence level.

[0180] For example, given the largest eigenvalue of the spatiotemporal covariance matrix, the product of the largest eigenvalue and 5% is determined as the second confidence level.

[0181] S407 determines the heart rate signal-to-noise ratio and multiple motion features based on the target frequency band of the micro-motion features.

[0182] Optionally, the target frequency band is the 0.5Hz to 2Hz band.

[0183] For example, the micro-motion features are subjected to phase unwinding to obtain the processed features, and then the processed features are filtered by a 0.5-2Hz bandpass filter to obtain the filtered features. Finally, the filtered features are subjected to CZT transformation to obtain the heart rate signal-to-noise ratio and multiple motion features.

[0184] Optionally, motion features are used to indicate the motion attributes displayed by the hand when the operation command is triggered. Motion features may include, but are not limited to, the rate of change of acceleration, the consistency of the direction of motion, and the smoothness of the motion trajectory, etc., without specific limitations.

[0185] S408 normalizes the heart rate signal-to-noise ratio to obtain the heart rate signal-to-noise ratio confidence level, and determines the continuity confidence level based on multiple motion features.

[0186] For example, the heart rate signal-to-noise ratio (HRNR) is normalized to obtain the HRNR confidence level; for instance, the product of the HRNR and 10% is determined as the HRNR confidence level. Furthermore, based on multiple motion features, a continuity confidence level is determined, which indicates the continuity confidence level of the trajectory that triggers the operation command.

[0187] When multiple motion features include the rate of change of acceleration, consistency of motion direction, and smoothness of motion trajectory, the continuity confidence of the trajectory triggering the joystick operation command is determined based on the rate of change of acceleration, consistency of motion direction, and smoothness of motion trajectory. Specifically, the priority weights of multiple motion features are obtained, and the continuity confidence is obtained based on the rate of change of acceleration, consistency of motion direction, smoothness of motion trajectory, and priority weights.

[0188] Optionally, the priority weights of multiple motion features are as follows: a weight of 0.4 for the rate of change of acceleration (or low Jerk), a weight of 0.3 for the consistency of motion direction, and a weight of 0.3 for the smoothness of the motion trajectory. Alternatively, the priority weights of multiple motion features can be as follows: a weight of 0.5 for the rate of change of acceleration, a weight of 0.3 for the consistency of motion direction, and a weight of 0.2 for the smoothness of the motion trajectory. The priority weights of multiple motion features can be determined according to the actual situation and are not specifically limited here.

[0189] S409 weights the heart rate signal-to-noise ratio confidence and the continuity confidence to obtain the third confidence level.

[0190] For example, given the confidence level of heart rate signal-to-noise ratio and the confidence level of continuity, the confidence level of heart rate signal-to-noise ratio and the confidence level of continuity are weighted to obtain a third confidence level.

[0191] S410, based on pressure characteristics, determines the degree of disorder in the pressure distribution on the joystick.

[0192] For example, the degree of disorder in the pressure distribution on the joystick is determined based on the pressure characteristics of the operation command. Optionally, the degree of disorder in the pressure distribution is used to indicate the degree of disorder in the pressure distribution on the joystick caused by the action that triggered the operation command; the degree of disorder in the pressure distribution can be represented by a pressure distribution entropy value.

[0193] S411, based on the pressure features, match them with the preset pressure features to obtain the feature matching degree.

[0194] For example, a feature matching degree is obtained by matching pressure features with preset pressure features. Optionally, the preset pressure features are used to indicate the pressure features corresponding to the target object, that is, to characterize the pressure distribution habits of the target object when holding the joystick. By matching the pressure features with the preset pressure features, the consistency between the pressure distribution and the preset pressure distribution can be determined, thereby determining the probability that the object triggering the operation command is the target object.

[0195] S412, based on the degree of disorder and feature matching, determine the fourth confidence level.

[0196] For example, a fourth confidence level is determined based on the degree of disorder and the feature matching degree. The degree of disorder is negatively correlated with the fourth confidence level, while the feature matching degree is positively correlated with the fourth confidence level. Since the degree of disorder is negatively correlated with the fourth confidence level, and the feature matching degree is positively correlated with the fourth confidence level, determining the fourth confidence level by comparing the degree of disorder in the pressure distribution with the feature matching degree corresponding to the pressure feature can improve the accuracy of identifying the object triggering the operation command as the target object, thereby improving the precise control of medical equipment.

[0197] S413, weight the third confidence level and the fourth confidence level to obtain the first confidence level.

[0198] For example, given the third and fourth confidence levels, the third and fourth confidence levels are weighted to obtain the first confidence level.

[0199] S414, When the first confidence level is greater than the first threshold and the second confidence level is greater than the second threshold, execute the operation instruction.

[0200] For example, given a first confidence level and a second confidence level, it is determined whether the first confidence level is greater than a first threshold and whether the second confidence level is greater than a second threshold. When the first confidence level is greater than the first threshold and the second confidence level is greater than the second threshold, it is determined to execute the operation instruction.

[0201] Optionally, the first threshold can be 0.7, 0.8, etc., without specific limitations.

[0202] Optionally, the second threshold can be 0.7, 0.8, etc., without specific limitations.

[0203] S415, if the first confidence level is less than or equal to the first threshold or the second confidence level is less than or equal to the second threshold, do not execute the operation instruction and output a warning message.

[0204] For example, when the first confidence level is less than or equal to the first threshold, or the second confidence level is less than or equal to the second threshold, it is determined that the operation instruction will not be executed, and a warning message is output. The warning message is used to warn that there may be a risk of accidental touch on the joystick of the medical device.

[0205] The above technical solution, since the first confidence level indicates the confidence that the object triggering the operation command is the target object, and the second confidence level indicates the confidence that the operation command is a valid operation, can analyze whether the operation command is the true intention of the target object through the first confidence level and the second confidence level. When the operation command is the true intention of the target object, the operation command is executed, which can reduce the risk of accidental touch of medical device operation, ensure accurate control of medical device, and thus improve the user experience.

[0206] It should be understood that the above examples are provided to help those skilled in the art understand the embodiments of this application, and are not intended to limit the embodiments of this application to the specific values ​​or scenarios illustrated. Those skilled in the art can obviously make various equivalent modifications or changes based on the above examples, and such modifications or changes also fall within the scope of the embodiments of this application.

[0207] The above text combined Figures 1 to 4 The control method for the medical device provided in the embodiments of this application is described in detail below; the following will be combined with Figure 5 and Figure 6 This application provides a detailed description of embodiments of the control device for a medical device. It should be understood that the control device for the medical device in these embodiments can execute the various methods described in the foregoing embodiments of this application. Specifically, the specific working processes of the various products described below can be referenced to the corresponding processes in the foregoing method embodiments.

[0208] Figure 5 This is a schematic diagram of the structure of a control device for a medical device provided in an embodiment of this application.

[0209] For example, the joystick of a medical device is equipped with a radar sensor and a piezoelectric sensor.

[0210] For example, such as Figure 5 As shown, the control device 500 of the medical device includes: The acquisition module 510 is used to acquire the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal in response to the operation command of the joystick. The digital baseband signal is used to indicate the signal acquired by the radar sensor, and the voltage signal is used to indicate the signal acquired by the piezoelectric sensor. The determination module 520 is used to determine a first confidence level and a second confidence level based on the micro-motion characteristics and pressure characteristics. The first confidence level is used to indicate the confidence level that the object triggering the operation command is the target object, and the second confidence level is used to indicate the confidence level that the operation command is a valid operation. The execution module 530 is used to determine whether to execute the operation instruction based on the first confidence level and the second confidence level.

[0211] Optionally, as an embodiment, the determining module 520 is specifically used for: Based on micro-motion features, the third confidence level of micro-motion features as human body features is determined; based on pressure features, the fourth confidence level of pressure features as target pressure features is determined, and the target pressure features are the pressure features corresponding to the target object; the third confidence level and the fourth confidence level are weighted to obtain the first confidence level.

[0212] Optionally, as an embodiment, the determining module 520 is specifically used for: Based on the target micro-motion features in the target frequency band of the micro-motion features, the heart rate signal-to-noise ratio and multiple motion features are determined; the heart rate signal-to-noise ratio is normalized to obtain the heart rate signal-to-noise ratio confidence level; based on multiple motion features, the continuity confidence level is determined, and the continuity confidence level is used to indicate the trajectory continuity confidence level of the trigger operation command; the heart rate signal-to-noise ratio confidence level and the continuity confidence level are weighted to obtain the third confidence level.

[0213] Optionally, as an embodiment, multiple motion characteristics include the rate of change of acceleration, the consistency of motion direction, and the smoothness of the motion trajectory; the determining module 520 is specifically used for: Obtain the priority weights of multiple motion features; based on the rate of change of acceleration, the consistency of motion direction, the smoothness of motion trajectory, and the priority weights, obtain the continuity confidence.

[0214] Optionally, as an embodiment, the determining module 520 is specifically used for: Based on pressure characteristics, the degree of disorder in the pressure distribution on the joystick is determined; based on the pressure characteristics, a feature matching degree is obtained by matching them with preset pressure characteristics, and the preset pressure characteristics are used to indicate the pressure characteristics corresponding to the target object; based on the degree of disorder and the feature matching degree, a fourth confidence level is determined, with the degree of disorder being negatively correlated with the fourth confidence level and the feature matching degree being positively correlated with the fourth confidence level.

[0215] Optionally, as an embodiment, the determining module 520 is specifically used for: Point cloud clustering is performed on the micro-motion features to obtain the acceleration features of hand joint movement; pressure gradient features are determined based on pressure features; and a second confidence level is determined based on the acceleration features and pressure gradient features.

[0216] Optionally, as an embodiment, the determining module 520 is specifically used for: Based on acceleration and pressure gradient features, a spatiotemporal covariance matrix is ​​constructed to indicate the degree of synchronization between acceleration and pressure gradient features. Based on the spatiotemporal covariance matrix, the largest eigenvalue is determined. The largest eigenvalue is numerically mapped to obtain the second confidence level.

[0217] Optionally, as an embodiment, the piezoelectric sensor includes multiple sensing units, and the voltage signal includes multiple voltage signals collected by the multiple sensing units; the acquisition module 510 is specifically used for: In response to the operation command of the joystick, it is determined whether there is an abnormal voltage signal among multiple voltage signals; when there is no abnormal voltage signal among multiple voltage signals, feature extraction is performed on the digital baseband signal and multiple voltage signals respectively to obtain micro-motion features and pressure features.

[0218] It should be noted that the control device 500 of the aforementioned medical equipment is embodied in the form of a functional unit. The term "module" here can be implemented in software and / or hardware, without specific limitations.

[0219] For example, a "module" can be a software program, a hardware circuit, or a combination of both that implements the above functions. The hardware circuit may include an application-specific integrated circuit (ASIC), electronic circuitry, a processor (e.g., a shared processor, a proprietary processor, or a group processor) and memory for executing one or more software or firmware programs, integrated logic circuitry, and / or other suitable components that support the described functions.

[0220] Therefore, the units of the various examples described in the embodiments of this application can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0221] Figure 6 This is a schematic diagram of the structure of a medical device provided in an embodiment of this application.

[0222] For example, such as Figure 6 As shown, the medical device 600 includes a memory 610 and a processor 620, wherein the memory 610 stores executable program code 630, and the processor 620 is used to call and execute the executable program code 630 to perform a control method for the medical device.

[0223] For example, memory 610 can be used to store programs related to the control method of the medical device provided in the embodiments of this application; processor 620 can call the programs related to the control method of the medical device stored in memory 610 to execute the control method of the medical device in the embodiments of this application; for example, in response to the operation command of the joystick, the micro-motion feature corresponding to the digital baseband signal and the pressure feature corresponding to the voltage signal are obtained, the digital baseband signal is used to indicate the signal collected by the radar sensor, and the voltage signal is used to indicate the signal collected by the piezoelectric sensor; based on the micro-motion feature and the pressure feature, a first confidence level and a second confidence level are determined, the first confidence level is used to indicate the confidence level that the object triggering the operation command is the target object, and the second confidence level is used to indicate the confidence level that the operation command is a valid operation; based on the first confidence level and the second confidence level, it is determined whether to execute the operation command.

[0224] This embodiment can divide functional modules according to the above method example. For example, each module can correspond to a separate functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware. It should be noted that the module division in this embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0225] It should be noted that all relevant content of each step involved in the above method embodiments can be referenced from the functional description of the corresponding functional module, and will not be repeated here.

[0226] When using integrated units, the device may include a processing module and a storage module. The processing module may be a processor or a controller that can implement or execute various exemplary logic blocks, modules, and circuits shown in conjunction with the disclosure of this application. The processor may also be a combination of functions that implement computing capabilities, such as a combination of one or more microprocessors, a combination of digital signal processing (DSP) and a microprocessor, etc., and the storage module may be a memory.

[0227] In addition, the device provided in the embodiments of this application may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute a control method for a medical device provided in the above embodiments.

[0228] This application also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, it causes the computer to execute the above-described related method steps to implement a control method for a medical device provided in the above embodiments. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, Digital Video Discs (DVDs), Compact Disc Read-Only Memory (CD-ROMs), microdrives, and magneto-optical disks, read-only memory (ROMs), random access memory (RAMs), erasable programmable read-only memory (EPROMs), electrically erasable programmable read-only memory (EEPROMs), dynamic random access memory (DRAMs), video random access memory (VRAMs), flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0229] This application also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to implement a medical device control method provided in the above embodiments.

[0230] The computer-readable storage medium, computer program product, or chip provided in this application are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0231] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0232] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0233] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for controlling a medical device, characterized in that, The joystick of the medical device is equipped with a radar sensor and a piezoelectric sensor; the method includes: In response to the operation command of the joystick, the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal are acquired. The digital baseband signal is used to indicate the signal acquired by the radar sensor, and the voltage signal is used to indicate the signal acquired by the piezoelectric sensor. Based on the micro-motion characteristics and the pressure characteristics, a first confidence level and a second confidence level are determined. The first confidence level is used to indicate the confidence level that the object triggering the operation command is the target object, and the second confidence level is used to indicate the confidence level that the operation command is a valid operation. Based on the first confidence level and the second confidence level, it is determined whether to execute the operation instruction.

2. The method according to claim 1, characterized in that, The determination of the first confidence level and the second confidence level based on the micro-motion feature and the pressure feature includes: Based on the micro-motion features, the third confidence level of the micro-motion features is determined to be a human body feature; Based on the pressure feature, the pressure feature is determined as the fourth confidence level of the target pressure feature, and the target pressure feature is the pressure feature corresponding to the target object; The third confidence level and the fourth confidence level are weighted to obtain the first confidence level.

3. The method according to claim 2, characterized in that, The third confidence level for determining the micro-motion feature as a human feature based on the micro-motion feature includes: Based on the target micro-motion features of the target frequency band in the micro-motion features, the heart rate signal-to-noise ratio and multiple motion features are determined; The heart rate signal-to-noise ratio is normalized to obtain the heart rate signal-to-noise ratio confidence level; Based on the aforementioned multiple motion features, a continuity confidence level is determined, which is used to indicate the continuity confidence level of the trajectory that triggers the operation command; The confidence level of the heart rate signal-to-noise ratio and the confidence level of continuity are weighted to obtain the third confidence level.

4. The method according to claim 3, characterized in that, The multiple motion characteristics include the rate of change of acceleration, the consistency of motion direction, and the smoothness of motion trajectory; The determination of continuity confidence based on the multiple motion features includes: Obtain the priority weights of the multiple motion features; The continuity confidence level is obtained based on the acceleration change rate, the consistency of the motion direction, the smoothness of the motion trajectory, and the priority weight.

5. The method according to claim 2, characterized in that, The fourth confidence level for determining the pressure feature as the target pressure feature based on the pressure feature includes: Based on the pressure characteristics, the degree of disorder in the pressure distribution on the joystick is determined; Based on the pressure feature, a feature matching degree is obtained by matching it with a preset pressure feature. The preset pressure feature is used to indicate the pressure feature corresponding to the target object. Based on the degree of disorder and the feature matching degree, the fourth confidence level is determined, wherein the degree of disorder is negatively correlated with the fourth confidence level, and the feature matching degree is positively correlated with the fourth confidence level.

6. The method according to claim 1, characterized in that, The determination of the first confidence level and the second confidence level based on the micro-motion feature and the pressure feature includes: Point cloud clustering is performed on the micro-motion features to obtain the acceleration features of hand joint motion; Based on the pressure characteristics, the pressure gradient characteristics are determined; The second confidence level is determined based on the acceleration characteristics and the pressure gradient characteristics.

7. The method according to claim 6, characterized in that, Determining the second confidence level based on the acceleration features and the pressure gradient features includes: Based on the acceleration features and the pressure gradient features, a spatiotemporal covariance matrix is ​​constructed, which is used to indicate the degree of synchronization between the acceleration features and the pressure gradient features. Based on the spatiotemporal covariance matrix, determine the largest eigenvalue; The second confidence level is obtained by numerically mapping the largest eigenvalue.

8. The method according to any one of claims 1 to 7, characterized in that, The piezoelectric sensor includes multiple sensing units, and the voltage signal includes multiple voltage signals collected by the multiple sensing units; The step of acquiring the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal in response to the operation command of the joystick includes: In response to an operation command on the joystick, determine whether there is an abnormal voltage signal among the plurality of voltage signals; When the abnormal voltage signal is not present among the plurality of voltage signals, feature extraction is performed on the digital baseband signal and the plurality of voltage signals respectively to obtain the micro-motion feature and the pressure feature.

9. A control device for a medical device, characterized in that, The joystick of the medical device is equipped with a radar sensor and a piezoelectric sensor; the device includes: The acquisition module is used to acquire, in response to the operation command of the joystick, the micro-motion characteristics corresponding to the digital baseband signal and the pressure characteristics corresponding to the voltage signal, wherein the digital baseband signal is used to indicate the signal acquired by the radar sensor and the voltage signal is used to indicate the signal acquired by the piezoelectric sensor; The determination module is used to determine a first confidence level and a second confidence level based on the micro-motion feature and the pressure feature. The first confidence level is used to indicate the confidence level that the object triggering the operation command is the target object, and the second confidence level is used to indicate the confidence level that the operation command is a valid operation. The execution module is used to determine whether to execute the operation instruction based on the first confidence level and the second confidence level.

10. A medical device, characterized in that, include: Memory, used to store executable program code; A processor is configured to call and run the executable program code from the memory, causing the medical device to perform the control method of the medical device as described in any one of claims 1 to 8.

Images