A control method of a self-service machine-controlled breathing switching device

Abstract

The application relates to the technical field of automation control, and discloses a control method of a self-service machine-controlled breathing switching device, which comprises the following steps: starting self-checking of wireless communication, remote control buttons and a voice prompt module, allowing remote switching after the starting, and otherwise notifying manual intervention. When running, the breathing frequency, the chest movement amplitude and the gas flow are continuously collected, and the normal baseline is compared to detect abnormalities. Meanwhile, the operation nodes, the image frame rate and the instrument operation state are fused to determine the key stage of the operation. According to the breathing deviation, the chest change, the tidal volume change and the node state, adaptive weights are dynamically generated, a switching index is calculated, the self-controlled breathing or the machine-controlled breathing is intelligently determined, and the machine-controlled breathing is forcedly switched when an abnormality occurs. After the determination, a command with a check is generated, is wirelessly sent to an anesthetic machine, is executed after being checked by the anesthetic machine, and feedback is returned, and if the execution fails, retry is triggered and a voice warning is given. The system continuously and dynamically samples, combines continuous abnormality safety processing, and forms a real-time monitoring closed loop.

Description

Technical Field

[0001] This invention relates to the field of automation control technology, specifically to a control method for a self-service machine-controlled breathing switching device. Background Technology

[0002] In interventional surgery, respiratory management is a core element in ensuring surgical safety and stabilizing patient vital signs. Interventional surgeries, especially high-precision procedures such as endovascular neurointerventions and cardiac stent implantation, typically require patients to maintain a stable or controllable respiratory state at specific stages to reduce imaging artifacts and avoid target vessel displacement, thereby ensuring the precision of the surgical procedure. Traditional respiratory control methods heavily rely on the anesthesiologist's experience and manual operation. This involves observing discrete vital sign parameters displayed on the monitor (such as respiratory rate and blood oxygen saturation) and visually assessing the patient's chest rise and fall, manually adjusting the anesthesia machine's operating mode (such as initiating mechanical ventilation) when abnormalities are detected. This approach not only has an inherent physiological delay in response but also suffers from significant subjective variability, making it difficult to quantify and standardize judgment criteria.

[0003] The complexity of the surgical environment further exacerbates the limitations of manual control. Instrument manipulation and image acquisition during critical surgical procedures such as DSA angiography, stent deployment, and balloon dilation drastically alter the requirements for patient respiratory stability. While the operator performs surgery outside the sterile area, the anesthesiologist needs to anticipate ventilation needs based on the surgical progress; however, a real-time, quantifiable information exchange mechanism between the two is often lacking. Although some existing assisted ventilation devices can provide basic spontaneous / mechanical switching, their decision-making logic is usually based solely on a single tidal volume or respiratory rate threshold. This fails to integrate multi-dimensional respiratory dynamic characteristics such as chest movement amplitude and real-time gas flow for comprehensive assessment, and also makes it difficult to perceive and adapt to dynamic risk changes brought about by critical surgical stages in real time. This results in a rigid switching mechanism that is easily triggered unnecessarily or delayed in response when truly needed.

[0004] Furthermore, in remote control operation scenarios, the reliability of command transmission, the self-checking capability of equipment status, and the overall closed-loop feedback of the system constitute technical shortcomings that urgently need improvement. Wireless control signals may attenuate or err due to factors such as electromagnetic shielding in the operating room and channel contention among multiple devices. Without a rigorous initialization self-check and data verification mechanism, serious risks of misoperation will arise. Current voice prompt systems mostly exist as isolated alarm modules, failing to be integrated with the entire closed-loop process of respiratory mode switching. They cannot provide clear status confirmation and anomaly guidance at each stage of initialization, decision-making, execution, and feedback, increasing the burden of intraoperative communication. Therefore, there is an urgent need for a highly reliable, multi-dimensional sensing, adaptive decision-making, and fully closed-loop verified intelligent machine-controlled respiratory switching solution to overcome the comprehensive deficiencies of traditional technologies in terms of accuracy, real-time performance, and safety. Summary of the Invention

[0005] This invention provides a control method for a self-service machine-controlled breathing switching device, which helps to solve the problems mentioned in the background art.

[0006] This invention provides the following technical solution: a control method for a self-service machine-controlled breathing switching device, comprising: By initializing and self-testing the wireless communication parameters, remote control button status, and voice prompt module, the breathing mode is switched after successful remote control initialization, and the staff is notified to intervene manually if the initialization fails. Collect patient respiratory rate, chest movement amplitude, and respiratory gas flow information to construct patient respiratory status data, and combine normal respiratory status to detect abnormal patient breathing and sensor status. By acquiring information on surgical nodes, the operating status of imaging equipment, and the operating status of instruments, the importance of key surgical nodes is calculated and smoothed to generate key node determination results. Based on the patient's respiratory rate deviation, changes in chest cavity movement, changes in tidal volume, and the status of key surgical nodes, adaptive weights are generated for each respiratory state, and the switching index between spontaneous breathing and mechanical breathing is calculated to determine the current breathing mode and switch to mechanical breathing mode when an abnormality is detected. Based on the determined breathing pattern, a corresponding control command is generated, and the control command is verified to form a data packet for transmission. The data packet is sent through the wireless communication module, and the legality of the received control command is verified. The anesthesia machine is controlled to execute the corresponding breathing mode. At the same time, execution feedback information is generated, and retry and voice prompts are given when execution fails or verification is abnormal. Based on the execution feedback information and the current breathing mode, generate corresponding voice prompts to provide voice feedback during the breathing mode switching process; Establish real-time status data, dynamically sample and update the patient's respiratory status in real time, continuously calculate the switching index between spontaneous breathing and mechanical breathing, and combine abnormal status to determine the dynamic breathing mode. At the same time, perform continuous abnormal safety handling to form a dynamic closed-loop control for breathing switching.

[0007] Optionally, the initialization and self-testing of wireless communication parameters, remote control button states, and voice prompt module, followed by switching the breathing mode upon successful remote control initialization and notifying staff for manual intervention upon initialization failure, includes: Construct a set of wireless communication parameters that includes multiple communication performance indicators, perform a communication self-test, and determine that the handshake is successful when all indicators meet the preset communication success conditions. Collect the working status of each button on the remote control and confirm that all button functions are normal. The volume, delay, and frequency characteristics of the voice prompt module are self-tested. When each characteristic meets the requirements of human ear recognition and synchronization, the voice self-test is deemed to have passed. The system performs a logical judgment based on three results: successful handshake, all buttons working properly, and voice self-test passed. If the overall result is true, the remote control initialization is successful and the breathing mode switch is executed; otherwise, the initialization fails and staff are notified to intervene manually.

[0008] Optionally, the step of collecting patient respiratory rate, chest movement amplitude, and respiratory gas flow information to construct patient respiratory status data, and combining this data with normal respiratory status to detect abnormal patient breathing and sensor status, includes: The respiratory airway pressure change waveform is obtained by a pressure sensor, the respiratory cycle is determined according to the time interval between adjacent peaks, and the real-time respiratory rate is calculated from the cycle. The displacement sensor is used to collect chest cavity displacement signals, and the maximum change in chest cavity during each respiratory cycle is extracted as the amplitude of chest cavity movement. Real-time respiratory gas flow is obtained by using a flow sensor, and the tidal volume of a single respiratory cycle is obtained by integrating the flow rate within a single respiratory cycle. The patient's respiratory state vector, composed of the real-time respiratory rate, chest movement amplitude, and tidal volume, is compared with the normal respiratory state vector, which is obtained in advance through statistics and composed of normal respiratory rate, normal chest movement amplitude, and normal tidal volume. If the deviation of any parameter exceeds the preset allowable deviation range, it is determined to be a respiratory abnormality or sensor malfunction.

[0009] Optionally, the step of calculating and smoothing the importance of key surgical nodes by acquiring surgical node information, imaging equipment operating status, and instrument operation status, and generating key node determination results, includes: Establish a set of surgical nodes and pre-define critical or non-critical status markers for each node; The system acquires the real-time frame rate of intraoperative imaging equipment and automatically generates instrument operation labels based on the key operation time intervals marked by the surgical operator. Based on node markers, image frame rate normalized to standard frame rate, and operation identifiers, a coupled normalization algorithm is used to determine the contribution of node markers, image frame rate, and instrument operation stability, and then generate node marker weight coefficients, image frame rate weight coefficients, and instrument operation state weight coefficients. The importance weight of each node is calculated using the aforementioned weighting coefficients. When the importance weight reaches or exceeds the preset critical node determination threshold, the node is initially determined to be a critical node; The weights of consecutive nodes are smoothed and averaged using a time window of a preset length, and then compared with the threshold again based on the smoothed weights to eliminate the influence of short-term fluctuations and obtain the final key node determination result.

[0010] Optionally, the step of generating adaptive weights for each respiratory state based on the patient's respiratory rate deviation, changes in chest cavity movement, changes in tidal volume, and the status of key surgical nodes, and calculating a switching index between spontaneous breathing and mechanical ventilation to determine the current respiratory mode, and switching to mechanical ventilation mode when an abnormality is detected, includes: The respiratory rate potential energy is calculated based on the degree of deviation of the real-time respiratory rate from the normal value. The thoracic motion potential energy is calculated based on the degree of deviation of the chest movement amplitude from the normal value and in combination with the intensity of the surgical node. The surgical node potential energy is calculated based on the intensity of the surgical node and in combination with the degree of deviation of the respiratory rate and the degree of deviation of the chest movement. The sum of the three is used as the total potential field. The respiratory rate weight, chest movement weight, and surgical node weight are calculated based on the respiratory rate potential energy, thoracic cavity motion potential energy, and surgical node potential energy, respectively. The tidal volume potential field is calculated based on the degree of deviation of the tidal volume from the normal value, and the tidal volume weight is determined by combining the respiratory rate potential field, the thoracic motion potential field, and the surgical node potential field. Real-time respiratory rate, chest movement amplitude, and tidal volume are normalized based on their normal values. The switching index between spontaneous breathing and mechanical breathing is calculated using the respiratory rate weight, chest movement weight, tidal volume weight, and key node markers. When the switching index is greater than the set threshold and no abnormal breathing sign appears, the current breathing mode is determined to be spontaneous breathing; otherwise, it is determined to be machine-controlled breathing mode. If signs of abnormal breathing are present, the breathing mode will be forcibly set to mechanical breathing.

[0011] Optionally, the step of generating corresponding control commands based on the determined breathing pattern, and forming a transmission data packet after data verification of the control commands, includes: Generate a corresponding preset binary instruction code based on the determined breathing pattern mapping; If an abnormal flag exists, an emergency stop instruction code is generated. Perform a cyclic redundancy check operation on the instruction code to obtain the check code; The instruction code, checksum, current timestamp, and key node determination flag are combined into a data packet to be sent.

[0012] Optionally, the step of sending the data packet via the wireless communication module, verifying the validity of the received control command, controlling the anesthesia machine to execute the corresponding breathing mode, generating execution feedback information, and retrying and providing voice prompts when execution fails or verification is abnormal includes: The data packet is sent using a wireless communication module with preset communication parameters, and the link quality is monitored. When the link quality meets the transmission requirements, the transmission is considered successful. At the receiving end, the cyclic redundancy check code is recalculated on the received instruction code and compared with the received check code. If the two match, the instruction is deemed valid. The breathing mode to be executed is determined according to the valid instruction code, and the corresponding operation is performed by the anesthesia machine; Generate a feedback vector containing the execution success or failure status, timestamp, key node flags, and actual execution mode; When an execution failure or invalid instruction verification occurs, an exception flag is set, and a reminder is issued to the operator via the voice prompt module, while the feedback vector is updated.

[0013] Optionally, the step of generating corresponding voice prompts based on execution feedback information and the current breathing mode to provide voice feedback during the breathing mode switching process includes: Based on the actual breathing pattern and whether the execution was successful or not in the feedback vector, the corresponding voice prompt text code is determined according to the preset mapping relationship; Based on the timestamp in the feedback vector and the preset voice delay parameters, calculate the actual playback time of the voice. Using the set volume and frequency configuration, output corresponding voice prompts at the playback time; Generate a voice playback feedback vector that includes the voice playback success status, playback time, and the encoded text of the played prompt.

[0014] Optionally, the establishment of real-time status data, dynamic sampling and real-time updating of the patient's respiratory status, continuous calculation of the spontaneous breathing and mechanical breathing switching index, and dynamic respiratory mode determination based on abnormal states, along with continuous abnormal safety handling, to form a dynamic closed-loop control for respiratory switching, includes: A real-time state vector is constructed, which includes the patient's real-time respiratory characteristics, chest cavity motion characteristics, tidal volume characteristics, key node markers, and abnormal markers, and is continuously updated according to a preset sampling frequency; The adaptive weights and normalized parameters are used to calculate the current spontaneous breathing and mechanical breathing switching index in real time. Based on whether the switching index is greater than a preset threshold and there is no abnormality indicator, the current breathing mode is dynamically determined to be spontaneous breathing or mechanical breathing. Based on the judgment result, generate and send instruction data packets, receive execution feedback from the anesthesia machine, and record logs for each execution, including timestamps, various status parameters, switching index, breathing mode, and execution status. Perform continuous anomaly safety handling. When the cumulative number of anomaly flags from multiple recent consecutive sampling times reaches the safety threshold, the breathing mode will be forcibly switched to machine-controlled breathing; otherwise, dynamic closed-loop control will be maintained.

[0015] The present invention has the following beneficial effects: 1. This method for controlling a self-service machine-controlled breathing switching device constructs a wireless communication vector containing parameters such as transmission power and channel frequency, designs a communication self-test function, sets a remote control button set and constructs a button checking algorithm, constructs a voice module parameter vector and self-test algorithm, and finally determines whether the initialization is successful through a comprehensive judgment function. If successful, the breathing mode is switched; if it fails, manual intervention is notified. The initialization and self-test mechanism proposed in this technical solution solves the problem of unreliable remote control in complex surgical environments. In interventional surgery, the anesthesia machine is usually placed outside the digital subtraction angiography room, with a thick protective wall separating the remote control and the anesthesia machine, resulting in extremely high risks of signal attenuation and interference. This method abandons the traditional single "power-on self-test" mode and innovatively constructs a multi-dimensional, parameterized joint self-test system. First, it transforms the evaluation of the wireless communication link from fuzzy empirical judgment to precise vector calculation. By quantifying specific indicators such as whether the transmission power reaches the minimum threshold for penetrating the protective door, whether the channel frequency is in the gap of the interference spectrum of numerous electronic devices in the operating room, and whether the link attenuation exceeds the tolerance, a "handshake success" flag is generated. This step ensures the absolute reliability of the command transmission channel at the physical level, rather than blindly sending commands. Furthermore, this method incorporates the two core aspects of human-computer interaction—button operation and voice feedback—into a self-testing closed loop, a major innovative breakthrough. Traditional remote controls only reveal their functionality when a button is pressed, while this solution proactively probes the electrical state of each button through a button checking algorithm. This allows for the early detection of hidden faults such as button jamming and poor contact, ensuring that every control command is accurately triggered and eliminating the fatal risk of "no response when pressed" in emergencies. Simultaneously, the self-test of the voice prompt module goes beyond simply "listening to a sound." It quantitatively analyzes whether the volume is clearly identifiable in the noise environment of an operating room, whether the prompt delay is synchronized with the operation, and whether the audio frequency is within the most sensitive range for the human ear. Ultimately, these three independent self-test results are fused through a multiplicative comprehensive judgment function, forming a rigorous "full-fact" judgment logic. This means that only when the communication link, physical buttons, and voice feedback are all in perfect condition does the system recognize the remote control initialization as successful and allow subsequent automatic breathing mode switching. Even the slightest flaw in any step will trigger a failure check and force staff to intervene manually. This "one-vote veto" initialization gating design completely intercepts the risk of system failure before automatic operation begins, establishing the highest safety threshold before the start of automated equipment in the operating room. This lays a solid foundation for the reliable execution of all subsequent automated processes and embodies the creative idea of ​​preventative safety design.

[0016] 2. This self-service machine-controlled breathing switching device control method constructs a breathing state vector by collecting the patient's respiratory rate, chest cavity movement amplitude, and respiratory gas flow rate, and compares it with a normal breathing state vector, setting thresholds to detect breathing abnormalities and sensor malfunctions. It describes how, by acquiring surgical node information, real-time imaging equipment frame rate, and instrument operation status, a designed coupled normalization algorithm dynamically calculates the weights of each factor, smooths the importance of nodes, and finally generates key node judgment results. This technical solution innovatively integrates the monitoring of the patient's internal physiological state with the perception of the external surgical process, constructing a two-dimensional breathing state assessment system with context-aware capabilities, solving the inherent problems of traditional ventilators or anesthesia machines having a single monitoring dimension and being detached from the surgical scenario. Its beneficial effects are mainly reflected in the depth and accuracy of perception and the intelligence of decision-making. First, the constructed patient breathing state detection is no longer a simple numerical threshold alarm. It constructs a three-dimensional breathing vector by simultaneously collecting three parameters—respiratory rate, chest cavity movement amplitude, and tidal volume—that describe the breathing state from three independent physical dimensions: time, space, and volume, respectively. Its innovation lies in the fact that the anomaly detection function does not judge a single parameter in isolation, but rather establishes a joint criterion. This design can keenly distinguish between "physiological abnormalities" and "sensor malfunctions." For example, when respiratory rate and tidal volume fluctuate drastically at the same frequency, but the chest cavity motion amplitude signal is abnormally stable, the system can determine that this is not a sudden change in the patient's condition, but most likely a loosening of the displacement sensor strap. This avoids triggering erroneous machine control switching due to sensor false alarms, greatly reducing the false alarm rate and improving the confidence of monitoring results. More importantly, the introduced determination of critical surgical nodes does not simply copy the pre-operative planned timetable, but dynamically defines the "criticality of the surgery" through real-time perception of the surgeon's behavior and the environmental state. Its core innovation lies in the design of a "node-image-instrument coupling normalization algorithm," which is like installing "eyes for understanding the surgical process" for the system. It no longer treats surgical nodes as static markers, but calculates the importance weight of the current moment in real time by analyzing changes in image frame rate and the micro-dynamics of hand movements during instrument operation. For example, when the surgeon is precisely positioning the stent, the real-time image frame rate reaches its peak to capture details, while the inter-frame displacement of the interventional instruments is extremely small, presenting a "quasi-static" state of fine-tuning. The algorithm in this solution can automatically capture this characteristic pattern and, through coupled normalization calculations, dynamically and significantly increase the weight of surgical nodes, even overriding the static node marker weights. Combined with time window smoothing, this mechanism can output a continuous, stable, and highly reflective critical node marker that accurately reflects the actual surgical tension. Ultimately, this dynamic surgical criticality information will serve as a core decision variable, inputting into the subsequent respiratory mode switching logic.This perception fusion architecture, which combines the patient's physiological state with the surgical procedure, endows the anesthesia safety control system with unprecedented "contextual awareness" capabilities, serving as the intelligent cornerstone for achieving precise, timely, and safe switching between autonomous and machine-controlled breathing.

[0017] 3. This control method for a self-service mechanical breathing switching device, based on respiratory rate deviation, changes in chest cavity movement, tidal volume changes, and the status of critical surgical nodes, generates adaptive weights by constructing a potential field function, calculates the switching index between spontaneous and mechanical breathing, determines the current breathing mode, and performs abnormal switching. Real-time status data is established, dynamically sampled, and the switching index is continuously calculated. Combined with abnormal states, continuous abnormal safety handling is performed, forming a dynamic closed-loop control for breathing switching. This technical solution proposes an adaptive decision-making algorithm based on a multi-factor competitive potential field, combined with a dynamic closed-loop control mechanism. This changes the rigid threshold judgment logic of traditional breathing mode switching, achieving a leap from "reflexive switching" to "intelligent predictive switching." Its beneficial effects are mainly reflected in the flexibility, adaptability, and system-level safety of decision-making. First, the designed adaptive weight generation mechanism is its core innovation. Instead of using fixed weights to integrate various indicators, it cleverly introduces the concept of "potential field" from physics. By constructing unique potential energy functions for respiratory rate deviation, changes in chest cavity movement, and the intensity of surgical nodes, the system can dynamically perceive the "decision urgency" of each factor. For example, when surgery reaches a critical stage, the potential energy of this surgical stage increases exponentially. Its weight dominates in the competition, causing the system to tend to switch to a safer machine-controlled mode even when the patient's respiratory parameters are slightly abnormal, creating an absolutely stable respiratory environment for delicate operations. Conversely, if only the respiratory rate fluctuates briefly, while the thoracic cavity motion potential energy is stable and the intensity of the surgical stage is low, the mathematical structure of the respiratory rate potential energy makes its growth relatively slow. The system will exhibit a certain degree of tolerance, avoiding excessive intervention. This "competitive" weight allocation makes the switching index a highly intelligent, dynamic threshold that understands the priority of the current situation. It is no longer a rigid numerical value, but a flexible decision curve that integrates anesthesia safety, surgical needs, and patient condition. Based on this, a closed loop of dynamic execution and safety assurance for this intelligent decision-making is constructed. Its innovation lies in building the entire judgment-execution-feedback process into a dynamic system that runs continuously on the timeline. The system samples and updates the patient's respiratory state and surgical progress at high frequency, continuously calculating the dynamic switching index. This ensures that the decision evolves in real time with the rapidly changing patient condition, rather than being an isolated judgment point. Even more ingeniously, the designed "continuous anomaly safety handling" mechanism, combined with the mechanism for immediate switching upon single anomaly trigger, forms a double layer of protection. By accumulating anomaly flags from the most recent moments, it only forcibly locks into the safe mode of mechanical ventilation when it confirms that the anomaly is persistent rather than a transient disturbance. This effectively filters out short-term signal spikes caused by the patient's momentary coughing or body movement, avoiding interference to the patient and the surgery caused by repeated switching. The system records the basis for each decision, the generated instructions, the execution results, and all current state variables as a complete log stream, forming a traceable and replayable complete control chain.This complete dynamic closed-loop design, from adaptive decision-making to continuous monitoring, from fault-tolerant processing to safety locking, elevates the automation and intelligence of anesthesia respiratory management to a new level. Its essence lies in the engineering implementation of the biomimetic control logic of "perception-thinking-action-confirmation". Attached Figure Description

[0018] Figure 1 This is a schematic diagram of the process of the present invention.

[0019] Figure 2 This is a diagram showing the dynamic changes in the adaptive potential field decision-making and switching index.

[0020] Figure 3 The SMI response surface plot for switching index.

[0021] Figure 4 A flowchart for the generation, sending, and execution verification of control instructions. Detailed Implementation

[0022] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] Example 1, see Figures 1 to 4 A control method for a self-service machine-controlled breathing switching device, comprising: By initializing and self-testing the wireless communication parameters, remote control button status, and voice prompt module, the breathing mode is switched after successful remote control initialization, and the staff is notified to intervene manually if the initialization fails. Collect patient respiratory rate, chest movement amplitude, and respiratory gas flow information to construct patient respiratory status data, and combine normal respiratory status to detect abnormal patient breathing and sensor status. By acquiring information on surgical nodes, the operating status of imaging equipment, and the operating status of instruments, the importance of key surgical nodes is calculated and smoothed to generate key node determination results. Based on the patient's respiratory rate deviation, changes in chest cavity movement, changes in tidal volume, and the status of key surgical nodes, adaptive weights are generated for each respiratory state, and the switching index between spontaneous breathing and mechanical breathing is calculated to determine the current breathing mode and switch to mechanical breathing mode when an abnormality is detected. Based on the determined breathing pattern, a corresponding control command is generated, and the control command is verified to form a data packet for transmission. The data packet is sent through the wireless communication module, and the legality of the received control command is verified. The anesthesia machine is controlled to execute the corresponding breathing mode. At the same time, execution feedback information is generated, and retry and voice prompts are given when execution fails or verification is abnormal. Based on the execution feedback information and the current breathing mode, generate corresponding voice prompts to provide voice feedback during the breathing mode switching process; Establish real-time status data, dynamically sample and update the patient's respiratory status in real time, continuously calculate the switching index between spontaneous breathing and mechanical breathing, and combine abnormal status to determine the dynamic breathing mode. At the same time, perform continuous abnormal safety handling to form a dynamic closed-loop control for breathing switching.

[0024] The process involves initializing and self-testing wireless communication parameters, remote control button states, and the voice prompt module. Upon successful remote control initialization, a breathing mode switch is executed. If initialization fails, staff are notified for manual intervention, including: Construct a set of wireless communication parameters that includes multiple communication performance indicators, perform a communication self-test, and determine that the handshake is successful when all indicators meet the preset communication success conditions. Collect the working status of each button on the remote control and confirm that all button functions are normal. The volume, delay, and frequency characteristics of the voice prompt module are self-tested. When each characteristic meets the requirements of human ear recognition and synchronization, the voice self-test is deemed to have passed. The system performs a logical judgment based on three results: successful handshake, all buttons working properly, and voice self-test passed. If the overall result is true, the remote control initialization is successful and the breathing mode switch is executed; otherwise, the initialization fails and staff are notified to intervene manually.

[0025] The process involves collecting patient respiratory rate, chest movement amplitude, and respiratory gas flow rate information to construct patient respiratory status data. This data, combined with information on normal respiratory status, is used to detect respiratory abnormalities and sensor status, including: The respiratory airway pressure change waveform is obtained by a pressure sensor, the respiratory cycle is determined according to the time interval between adjacent peaks, and the real-time respiratory rate is calculated from the cycle. The displacement sensor is used to collect chest cavity displacement signals, and the maximum change in chest cavity during each respiratory cycle is extracted as the amplitude of chest cavity movement. Real-time respiratory gas flow is obtained by using a flow sensor, and the tidal volume of a single respiratory cycle is obtained by integrating the flow rate within a single respiratory cycle. The patient's respiratory state vector, composed of the real-time respiratory rate, chest movement amplitude, and tidal volume, is compared with the normal respiratory state vector, which is obtained in advance through statistics and composed of normal respiratory rate, normal chest movement amplitude, and normal tidal volume. If the deviation of any parameter exceeds the preset allowable deviation range, it is determined to be a respiratory abnormality or sensor malfunction.

[0026] The process involves acquiring surgical node information, imaging equipment operating status, and instrument operation status to calculate and smooth the importance of key surgical nodes, generating key node determination results, including: Establish a set of surgical nodes and pre-define critical or non-critical status markers for each node; The system acquires the real-time frame rate of intraoperative imaging equipment and automatically generates instrument operation labels based on the key operation time intervals marked by the surgical operator. Based on node markers, image frame rate normalized to standard frame rate, and operation identifiers, a coupled normalization algorithm is used to determine the contribution of node markers, image frame rate, and instrument operation stability, and then generate node marker weight coefficients, image frame rate weight coefficients, and instrument operation state weight coefficients. The importance weight of each node is calculated using the aforementioned weighting coefficients. When the importance weight reaches or exceeds the preset critical node determination threshold, the node is initially determined to be a critical node; The weights of consecutive nodes are smoothed and averaged using a time window of a preset length, and then compared with the threshold again based on the smoothed weights to eliminate the influence of short-term fluctuations and obtain the final key node determination result.

[0027] The process involves generating adaptive weights for each respiratory state based on deviations in the patient's respiratory rate, changes in chest movement, changes in tidal volume, and the status of critical surgical points. It also calculates a switching index between spontaneous and mechanical ventilation to determine the current respiratory mode and switches to mechanical ventilation mode when an abnormality is detected. This includes: The respiratory rate potential energy is calculated based on the degree of deviation of the real-time respiratory rate from the normal value. The thoracic motion potential energy is calculated based on the degree of deviation of the chest movement amplitude from the normal value and in combination with the intensity of the surgical node. The surgical node potential energy is calculated based on the intensity of the surgical node and in combination with the degree of deviation of the respiratory rate and the degree of deviation of the chest movement. The sum of the three is used as the total potential field. The respiratory rate weight, chest movement weight, and surgical node weight are calculated based on the respiratory rate potential energy, thoracic cavity motion potential energy, and surgical node potential energy, respectively. The tidal volume potential field is calculated based on the degree of deviation of the tidal volume from the normal value, and the tidal volume weight is determined by combining the respiratory rate potential field, the thoracic motion potential field, and the surgical node potential field. Real-time respiratory rate, chest movement amplitude, and tidal volume are normalized based on their normal values. The switching index between spontaneous breathing and mechanical breathing is calculated using the respiratory rate weight, chest movement weight, tidal volume weight, and key node markers. When the switching index is greater than the set threshold and no abnormal breathing sign appears, the current breathing mode is determined to be spontaneous breathing; otherwise, it is determined to be machine-controlled breathing mode. If signs of abnormal breathing are present, the breathing mode will be forcibly set to mechanical breathing.

[0028] The process of generating corresponding control commands based on the determined breathing pattern, and then verifying the control commands to form a data packet for transmission, includes: Generate a corresponding preset binary instruction code based on the determined breathing pattern mapping; If an abnormal flag exists, an emergency stop instruction code is generated. Perform a cyclic redundancy check operation on the instruction code to obtain the check code; The instruction code, checksum, current timestamp, and key node determination flag are combined into a data packet to be sent.

[0029] The process involves sending the data packet via a wireless communication module, validating the received control commands, controlling the anesthesia machine to execute the corresponding breathing mode, generating execution feedback information, and providing retry and voice prompts in case of execution failure or verification abnormalities. The data packet is sent using a wireless communication module with preset communication parameters, and the link quality is monitored. When the link quality meets the transmission requirements, the transmission is considered successful. At the receiving end, the cyclic redundancy check code is recalculated on the received instruction code and compared with the received check code. If the two match, the instruction is deemed valid. The breathing mode to be executed is determined according to the valid instruction code, and the corresponding operation is performed by the anesthesia machine; Generate a feedback vector containing the execution success or failure status, timestamp, key node flags, and actual execution mode; When an execution failure or invalid instruction verification occurs, an exception flag is set, and a reminder is issued to the operator via the voice prompt module, while the feedback vector is updated.

[0030] The step of generating corresponding voice prompts based on execution feedback information and the current breathing mode to provide voice feedback during the breathing mode switching process includes: Based on the actual breathing pattern and whether the execution was successful or not in the feedback vector, the corresponding voice prompt text code is determined according to the preset mapping relationship; Based on the timestamp in the feedback vector and the preset voice delay parameters, calculate the actual playback time of the voice. Using the set volume and frequency configuration, output corresponding voice prompts at the playback time; Generate a voice playback feedback vector that includes the voice playback success status, playback time, and the encoded text of the played prompt.

[0031] The establishment of real-time status data, dynamic sampling and real-time updating of the patient's respiratory status, continuous calculation of the spontaneous breathing and mechanical ventilation switching index, and dynamic respiratory mode determination based on abnormal states, along with continuous abnormal safety handling, to form a dynamic closed-loop control for respiratory switching, includes: A real-time state vector is constructed, which includes the patient's real-time respiratory characteristics, chest cavity motion characteristics, tidal volume characteristics, key node markers, and abnormal markers, and is continuously updated according to a preset sampling frequency; The adaptive weights and normalized parameters are used to calculate the current spontaneous breathing and mechanical breathing switching index in real time. Based on whether the switching index is greater than a preset threshold and there is no abnormality indicator, the current breathing mode is dynamically determined to be spontaneous breathing or mechanical breathing. Based on the judgment result, generate and send instruction data packets, receive execution feedback from the anesthesia machine, and record logs for each execution, including timestamps, various status parameters, switching index, breathing mode, and execution status. Perform continuous anomaly safety handling. When the cumulative number of anomaly flags from multiple recent consecutive sampling times reaches the safety threshold, the breathing mode will be forcibly switched to machine-controlled breathing; otherwise, dynamic closed-loop control will be maintained.

[0032] Example 2: A control method for a self-service machine-controlled breathing switching device, comprising: By initializing and self-testing the wireless communication parameters, remote control button status, and voice prompt module, the breathing mode is switched after successful remote control initialization, and the staff is notified to intervene manually if the initialization fails. Collect patient respiratory rate, chest movement amplitude, and respiratory gas flow information to construct patient respiratory status data, and combine normal respiratory status to detect abnormal patient breathing and sensor status. By acquiring information on surgical nodes, the operating status of imaging equipment, and the operating status of instruments, the importance of key surgical nodes is calculated and smoothed to generate key node determination results. Based on the patient's respiratory rate deviation, changes in chest cavity movement, changes in tidal volume, and the status of key surgical nodes, adaptive weights are generated for each respiratory state, and the switching index between spontaneous breathing and mechanical breathing is calculated to determine the current breathing mode and switch to mechanical breathing mode when an abnormality is detected. Based on the determined breathing pattern, a corresponding control command is generated, and the control command is verified to form a data packet for transmission. The data packet is sent through the wireless communication module, and the legality of the received control command is verified. The anesthesia machine is controlled to execute the corresponding breathing mode. At the same time, execution feedback information is generated, and retry and voice prompts are given when execution fails or verification is abnormal. Based on the execution feedback information and the current breathing mode, generate corresponding voice prompts to provide voice feedback during the breathing mode switching process; Establish real-time status data, dynamically sample and update the patient's respiratory status in real time, continuously calculate the switching index between spontaneous breathing and mechanical breathing, and combine abnormal status to determine the dynamic breathing mode. At the same time, perform continuous abnormal safety handling to form a dynamic closed-loop control for breathing switching.

[0033] The process involves initializing and self-testing wireless communication parameters, remote control button states, and the voice prompt module. Upon successful remote control initialization, a breathing mode switch is executed. If initialization fails, staff are notified for manual intervention, including: Constructing wireless communication parameter vectors ; in, Indicates the transmission power. Indicate the channel frequency to ensure no interference with other devices. This represents a data check bit, ensuring data integrity. This indicates the link attenuation threshold; if this value is exceeded, the power needs to be readjusted. This indicates a signal response delay, ensuring real-time command transmission; Construct a communication self-test function Ensure that the remote control signal can cover the outdoor operating location of the DSA: ; in, The handshake is indicated by a 1 for success and a 0 for failure. This represents the minimum transmit power threshold. Represents the set of allowed channels. Indicates the minimum number of parity bits. Indicates the maximum link attenuation. Indicates the maximum response delay; Set remote control button set , This indicates the m-th button; Collect each button Status: ; in, Indicates button Status identifier code; A button checking algorithm is built to ensure that each remote control button can correctly trigger the corresponding breathing switching function: ; in, This indicates the self-test result of the remote control button status; A value of 1 indicates that all buttons are functioning correctly, while a value of 0 indicates that at least one button is malfunctioning. Set the voice module parameter vector ; in, Indicates volume level. This indicates a delay in the prompt, ensuring synchronization with the button press. Indicates the range of audio frequencies; Build a voice self-checking algorithm to ensure that every key press is accompanied by voice confirmation: ; in, This indicates the self-test results of the voice prompt module. Indicates the minimum identifiable volume. Indicates the maximum voice delay. Indicates the range of frequencies audible to the human ear; Construct the RCI synthesis judgment function: ; in, This indicates the comprehensive judgment result of the remote control initialization status; like A value of 1 indicates that the remote control initialization was successful, and then the breathing mode switching will proceed; like A value of 0 indicates initialization failure, in which case staff will be notified to intervene manually.

[0034] By constructing a wireless communication vector containing parameters such as transmit power and channel frequency, a communication self-test function is designed; a remote control button set is set and a button checking algorithm is constructed; a voice module parameter vector and self-test algorithm are constructed; and finally, a comprehensive judgment function determines whether the initialization is successful. If successful, a breathing mode switch is executed; if unsuccessful, manual intervention is notified. The initialization and self-test mechanism proposed in this technical solution solves the problem of unreliable remote control in complex surgical environments. In interventional surgery, the anesthesia machine is usually placed outside the digital subtraction angiography room, with a thick protective wall separating the remote control and the anesthesia machine, resulting in extremely high risks of signal attenuation and interference. This method abandons the traditional single "power-on self-test" mode and innovatively constructs a multi-dimensional, parameterized joint self-test system. First, it transforms the evaluation of the wireless communication link from fuzzy empirical judgment to precise vector calculation. By quantifying specific indicators such as whether the transmit power reaches the minimum threshold for penetrating the protective door, whether the channel frequency is in the gap of the interference spectrum of numerous electronic devices in the operating room, and whether the link attenuation exceeds the tolerance, a "handshake success" flag is generated. This step ensures the absolute reliability of the command transmission channel at the physical level, rather than blindly sending commands. Furthermore, this method incorporates two core aspects of human-computer interaction—button operation and voice feedback—into a self-checking closed loop, a major innovative breakthrough. Traditional remote controls only reveal their functionality upon pressing a button, while this solution proactively probes the electrical state of each button through a button-checking algorithm. This allows for the early detection of hidden faults such as button jamming and poor contact, ensuring accurate triggering of every control command and eliminating the fatal risk of "no response" in emergencies. Simultaneously, the self-check of the voice prompt module goes beyond simply "listening to a sound." It quantitatively analyzes whether the volume is clearly identifiable in the noisy operating room environment, whether the prompt delay is synchronized with the operation, and whether the audio frequency is within the most sensitive range for the human ear. Ultimately, these three independent self-check results are fused through a multiplicative comprehensive judgment function, forming a rigorous "full-fact" judgment logic. This means that only when the communication link, physical buttons, and voice feedback are all in perfect condition does the system consider the remote control initialization successful and allow subsequent automatic breathing mode switching. Even a minor flaw in any of these components triggers a failure judgment and forces manual intervention from staff. This "one-vote veto" initialization gating design completely intercepts the risk of system failure before automatic operation starts, establishing the highest safety threshold before the start of automated equipment in the operating room, laying a solid foundation for the reliable execution of all subsequent automated processes, and reflecting the creative idea of ​​preventive safety design.

[0035] The process involves collecting patient respiratory rate, chest movement amplitude, and respiratory gas flow rate information to construct patient respiratory status data. This data, combined with information on normal respiratory status, is used to detect respiratory abnormalities and sensor status, including: Initialize the breathing sensor: ; in, This represents the state vector of the respiratory sensor. Indicates real-time respiratory rate, This indicates the magnitude of changes in respiratory airway pressure. Indicates the sensor sampling frequency; Calculate the respiratory cycle based on the pressure change waveform. : ; in, For the first The timing of the second respiratory peak; The real-time respiratory rate is specifically: ; Using a displacement sensor to collect chest cavity motion: ; in, This represents the vector of thoracic cavity motion state. Indicates the range of motion of the chest cavity. This indicates the reference value for the tightness and length of the chest strap. Indicates the sampling frequency; Displacement signal output from displacement sensor Extract the maximum chest cavity change during each breath, i.e., the amplitude of chest cavity movement: ; Gas flow rate was collected using a flow sensor from an anesthesia machine. ; in, Represents the respiratory gas flow vector. Indicates real-time respiratory gas flow rate. Indicates the sampling frequency. Tidal volume, which is the volume of gas inhaled per respiratory cycle, can be calculated by integrating the flow rate. , and These are the start and end times of a single respiratory cycle; Constructing a unified respiratory state vector ; Construct a normal respiratory state vector through data statistics: ; in, Represents a normal breathing state vector. This indicates a normal respiratory rate. This indicates the normal chest cavity amplitude. This indicates normal tidal volume; Construct an anomaly detection formula to detect abnormal breathing or sensor malfunction: ; in, This indicates an abnormality; 1 indicates an abnormality, and 0 indicates a normality. These represent the respiratory rate threshold, chest movement amplitude threshold, and tidal volume threshold, respectively, determined by the statistical sample.

[0036] The process involves acquiring surgical node information, imaging equipment operating status, and instrument operation status to calculate and smooth the importance of key surgical nodes, generating key node determination results, including: Establish a set of surgical nodes ; in, Indicates the first Each node has a status: 0 for non-critical, 1 for critical. This indicates the total number of surgical nodes, set according to the preoperative surgical plan. Obtain the frame rate of the imaging equipment and the surgical operation state vector. ; in, Indicates the real-time video frame rate. Operation timestamp, Instrument operation indicators can be automatically generated by the operator marking key operation stages in real time during surgery. Specifically: ;· Set node weights: ; in, Indicates the first The importance weight of each node Indicates the node label weight coefficient. This represents the image frame rate weighting coefficient. This represents the weighting coefficient of the operation state. Indicates the standard frame rate, used for normalization; The design node-image-instrument coupling normalization algorithm calculates the above weight coefficients as follows: ; ; ; in, Indicates the node's contribution. This indicates the contribution of image frame rate. The contribution of instrument operation stability is expressed as follows: , , ; in, This indicates the number of times key nodes are marked within the current statistics window. This indicates the total number of records for all nodes within the current statistics window. Indicates the current DSA image frame rate. This indicates the maximum image frame rate of the DSA device. Indicates the coordinates of the device in the current frame. Indicates the coordinates of the instrument in the previous frame. This represents the instrument micro-displacement threshold, obtained from preoperative calibration. This indicates the current inter-frame displacement of the instrument, specifically: ; Construct a node determination formula to generate key node markers: ; in, This indicates a critical node; 1 indicates critical, 0 indicates non-critical. The threshold for determining critical nodes is obtained from statistical analysis of historical surgical data. Use length is Weight smoothing is applied to consecutive nodes within a time window: ; in, This represents the smoothed node weights; The smoothed node weights are then used to determine the outcome, eliminating the impact of short-term fluctuations on the decision. .

[0037] This paper describes a respiratory state vector constructed by collecting patient respiratory rate, chest cavity movement amplitude, and respiratory gas flow rate. This vector is compared with a normal respiratory state vector, and thresholds are set to detect respiratory abnormalities and sensor malfunctions. The paper describes how, by acquiring surgical node information, real-time imaging equipment frame rate, and instrument operation status, a designed coupled normalization algorithm dynamically calculates the weights of each factor and smooths the node importance, ultimately generating a key node judgment result. This innovative technical solution deeply and organically integrates the monitoring of the patient's internal physiological state with the perception of the external surgical process, constructing a two-dimensional respiratory state assessment system with context-aware capabilities. This solves the inherent problems of traditional ventilator or anesthesia machine monitoring being singular in dimension and detached from the surgical scenario. Its beneficial effects are mainly reflected in the depth and accuracy of perception and the intelligence of decision-making. First, the constructed patient respiratory state detection is no longer a simple numerical threshold alarm. It constructs a three-dimensional respiratory vector by simultaneously collecting three parameters—respiratory rate, chest cavity movement amplitude, and tidal volume—that describe the respiratory state from three independent physical dimensions: time, space, and volume, respectively. Its innovation lies in the fact that the anomaly detection function does not judge a single parameter in isolation, but rather establishes a joint criterion. This design can keenly distinguish between "physiological abnormalities" and "sensor malfunctions." For example, when respiratory rate and tidal volume fluctuate dramatically at the same frequency, but the chest movement amplitude signal remains abnormally stable, the system can determine that this is not a sudden change in the patient's condition, but most likely a loosening of the displacement sensor strap. This avoids triggering erroneous machine-controlled switching due to sensor false alarms, greatly reducing the false alarm rate and improving the confidence of monitoring results. More importantly, the introduced determination of critical surgical nodes does not simply follow the pre-operative timetable, but dynamically defines the "criticality of the surgery" through real-time perception of the surgeon's behavior and the environmental state. Its core innovation lies in the design of a "node-image-instrument coupling normalization algorithm," which acts like "eyes that understand the surgical process" for the system. It no longer treats surgical nodes as static markers, but calculates the importance weight of the current moment in real time by analyzing changes in image frame rate and the micro-dynamics of hand movements during instrument operation. For example, when the surgeon is precisely positioning the stent, the real-time image frame rate will reach its peak to capture details, while the inter-frame displacement of the interventional instruments will be extremely small, presenting a "quasi-static" state of fine operation. The algorithm in this scheme can automatically capture this feature pattern and dynamically and significantly increase the weight of surgical nodes through coupled normalization calculations, even overriding the static node label weights. Combined with time window smoothing processing, this mechanism can output a continuous, stable, and highly reflective critical node marker that accurately reflects the actual surgical tension. Ultimately, this dynamic surgical criticality information will serve as a core decision variable, inputting into the subsequent breathing mode switching logic.This perception fusion architecture, which combines the patient's physiological state with the surgical procedure, endows the anesthesia safety control system with unprecedented "contextual awareness" capabilities, serving as the intelligent cornerstone for achieving precise, timely, and safe switching between autonomous and machine-controlled breathing.

[0038] The process involves generating adaptive weights for each respiratory state based on deviations in the patient's respiratory rate, changes in chest movement, changes in tidal volume, and the status of critical surgical points. It also calculates a switching index between spontaneous and mechanical ventilation to determine the current respiratory mode and switches to mechanical ventilation mode when an abnormality is detected. This includes: Set respiratory rate weight Chest movement weight and surgical node weight The specific process is as follows: Calculate respiratory rate potential energy; when respiratory rate is abnormal but chest rise and fall are not yet obvious, increase sensitivity to respiratory rate abnormalities: ; in, Represents the potential energy of respiratory rate. This indicates the deviation of respiratory rate, specifically: , This indicates the deviation of chest cavity movement, specifically: , To prevent extremely small positive numbers with a denominator of 0; Calculate the thoracic cavity kinetic energy: ; in, It represents the kinetic potential energy of the thoracic cavity. This indicates the intensity of the surgical node and can be preset based on historical data. When at critical surgical points and when chest cavity movement is significant, the decision-making influence of chest cavity movement factors is significantly increased. Calculate the potential energy at the surgical node: ; in, Represents the potential energy at the surgical node; When entering critical stages such as stent release and positioning, even if respiratory abnormalities are not obvious, the tendency for mechanically regulated breathing can be enhanced. Constructing the total amount of competition in the potential field : ; Generate adaptive weights: , , ; To standardize the units of measurement, the patient's respiratory status was normalized: ; in, Indicates normalized respiratory rate. This represents the normalized chest cavity amplitude. Indicates normalized tidal volume; Construct a switching function and calculate the autonomous-machine-controlled switching index: ; in, This represents the autonomous-machine-controlled switching index, with a value range of 0 to 1. The tidal volume weight is represented by the following calculation process: Calculate tidal volume deviation : ; Construct a potential field calculation formula to calculate the tidal volume potential field. : ; Used to indicate the intensity of the impact of abnormal tidal volume on respiratory switching; Calculate tidal volume weight : ; in, The potential field representing the respiratory rate is calculated using the potential field calculation formula. This represents the potential field of the thoracic cavity, calculated using the potential field calculation formula. The potential field is calculated using the formula. Respiratory pattern is determined based on the autonomous-mechanical switching index: ; in, Indicates breathing pattern; like This will trigger the security mechanism and set... .

[0039] The process of generating corresponding control commands based on the determined breathing pattern, and then verifying the control commands to form a data packet for transmission, includes: To ensure the anesthesia machine can recognize the breathing pattern, the corresponding instruction code is defined: ; in, This represents a binary instruction code, which is sent from the remote control to the anesthesia machine. Each code uniquely corresponds to a pattern to ensure that the parameters are not repeated. Will The mapping is to instructions, specifically: when At that time, ; when At that time, ; When the abnormal flag At that time, Emergency mode; To avoid accidental operations, a checksum is generated using CRC verification. ; in, This represents a 16-bit checksum used to verify the integrity of the instruction. This refers to the CRC function, which performs check calculations on the instruction code, as follows: Define input variables: ; in, The first instruction code Bit binary number, Indicates the length of the instruction code; Define the initial CRC register: ; in, Indicates a 16-bit register. Initial value; Generating polynomials: ; in, This represents the generating function, and x represents the shift variable, used to represent the shift level of binary data in the register. Extend each bit of the instruction code to 16 bits for processing: ; in, This represents the extended instruction code; For each Perform the following operations: ; in, This indicates a bitwise XOR operation. This indicates a left shift of 8 bits, placing the current bit in the high-order bit of the register. Perform custom polynomial processing on each bit of the register: ; in, Indicates the most significant bit of the register. This represents 16-bit generator polynomial coefficients, corresponding to , This indicates a left shift by one bit; Output the CRC after calculation: ; Generate the final data packet to be sent .

[0040] The process involves sending the data packet via a wireless communication module, validating the received control commands, controlling the anesthesia machine to execute the corresponding breathing mode, generating execution feedback information, and providing retry and voice prompts in case of execution failure or verification abnormalities. Send signal vector via wireless communication module ; Construct a sending determination algorithm to determine in real time whether the command has been successfully sent: ; in, This indicates a sending determination flag: 1 for success, 0 for failure. Perform CRC check: ; in, This indicates the received breathing instruction code. This indicates the received verification code. This indicates the validity of the instruction: 1 for valid, 0 for invalid. Based on the received instructions, the anesthesia machine performs a breathing switch: ; in, Indicates the actual execution mode; Generate feedback vector ; in, Indicates the execution status: 1 for success, 0 for failure; If the execution fails or the CRC check is invalid, the details are as follows: ; in, This indicates an execution exception flag: 1 for exception, 0 for normal. like It issues a voice prompt to the operator; Update feedback vector .

[0041] The step of generating corresponding voice prompts based on execution feedback information and the current breathing mode to provide voice feedback during the breathing mode switching process includes: Constructing speech parameter vectors ; in, Indicates the volume of the voice message. Indicates the speech delay time. Indicates audio frequency. The text encoding representing the voice prompts corresponding to different operations; Based on the feedback vector Mapped voice prompts: ; The mapping rules are as follows: ; Design a formula for synchronizing voice playback time: ; in, Indicates the actual playback time of the audio; Volume and frequency settings: ; Design a voice playback function: ; in, Indicates the voice playback status: 1 for success, 0 for failure. Indicates volume and frequency settings; Generate audio playback feedback vectors .

[0042] The establishment of real-time status data, dynamic sampling and real-time updating of the patient's respiratory status, continuous calculation of the spontaneous breathing and mechanical ventilation switching index, and dynamic respiratory mode determination based on abnormal states, along with continuous abnormal safety handling, to form a dynamic closed-loop control for respiratory switching, includes: Establish real-time state vector ; Set data sampling frequency ; Calculate the dynamic autonomous-machine-controlled switching index using the switching function: ; in, Indicates a dynamically recursive switching index; Setting dynamic thresholds for judgment: ; in, Indicates the current breathing pattern; Generate the instruction packet for the current moment. ; Send command packet And receive the current feedback vector ; This forms a dynamic closed-loop control, ensuring that each breathing switch is executed accurately and confirmed in real time. Record each dynamic switch: ; in, This indicates system log records; Perform continuous exception handling: ; in, This indicates the current system's safe operating mode. This represents the sum of the anomaly flags from the most recent three moments.

[0043] Based on respiratory rate deviation, changes in chest cavity movement, tidal volume changes, and the status of critical surgical nodes, an adaptive weight is generated by constructing a potential field function to calculate the switching index between spontaneous and mechanically induced breathing, thereby determining the current breathing mode and initiating abnormal switching. Real-time status data is established, dynamically sampled, and the switching index is continuously calculated. Combined with abnormal states, continuous abnormal safety handling is performed, forming a dynamic closed-loop control for breathing switching. This technical solution proposes an adaptive decision-making algorithm based on a multi-factor competitive potential field, combined with a dynamic closed-loop control mechanism. This changes the rigid threshold judgment logic of traditional breathing mode switching, achieving a leap from "reflexive switching" to "intelligent predictive switching." Its beneficial effects are mainly reflected in the flexibility, adaptability, and system-level safety of decision-making. First, the designed adaptive weight generation mechanism is its core innovation. Instead of using fixed weights to integrate various indicators, it cleverly introduces the concept of "potential field" from physics. By constructing unique potential energy functions for respiratory rate deviation, changes in chest cavity movement, and the intensity of surgical nodes, the system can dynamically perceive the "decision urgency" of each factor. For example, when surgery reaches a critical stage, the potential energy of this surgical stage increases exponentially. Its weight dominates in the competition, causing the system to tend to switch to a safer machine-controlled mode even when the patient's respiratory parameters are slightly abnormal, creating an absolutely stable respiratory environment for delicate operations. Conversely, if only the respiratory rate fluctuates briefly, while the thoracic cavity motion potential energy is stable and the intensity of the surgical stage is low, the mathematical structure of the respiratory rate potential energy makes its growth relatively slow. The system will exhibit a certain degree of tolerance, avoiding excessive intervention. This "competitive" weight allocation makes the switching index a highly intelligent, dynamic threshold that understands the priority of the current situation. It is no longer a rigid numerical value, but a flexible decision curve that integrates anesthesia safety, surgical needs, and patient condition. Based on this, a closed loop of dynamic execution and safety assurance for this intelligent decision-making is constructed. Its innovation lies in building the entire judgment-execution-feedback process into a dynamic system that runs continuously on the timeline. The system samples and updates the patient's respiratory state and surgical progress at high frequency, continuously calculating the dynamic switching index. This ensures that the decision evolves in real time with the rapidly changing patient condition, rather than being an isolated judgment point. Even more ingeniously, the designed "continuous anomaly safety handling" mechanism, combined with the mechanism for immediate switching upon single anomaly trigger, forms a double layer of protection. By accumulating anomaly flags from the most recent moments, it only forcibly locks into the safe mode of mechanical ventilation when it confirms that the anomaly is persistent rather than a transient disturbance. This effectively filters out short-term signal spikes caused by the patient's momentary coughing or body movement, avoiding interference to the patient and the surgery caused by repeated switching. The system records the basis for each decision, the generated instructions, the execution results, and all current state variables as a complete log stream, forming a traceable and replayable complete control chain.This complete dynamic closed-loop design, from adaptive decision-making to continuous monitoring, from fault-tolerant processing to safety locking, elevates the automation and intelligence of anesthesia respiratory management to a new level. Its essence lies in the engineering implementation of the biomimetic control logic of "perception-thinking-action-confirmation".

[0044] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0045] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A control method for a self-service machine-controlled breathing switching device, characterized in that, include: By initializing and self-testing the wireless communication parameters, remote control button status, and voice prompt module, the breathing mode is switched after successful remote control initialization, and the staff is notified to intervene manually if the initialization fails. Collect patient respiratory rate, chest movement amplitude, and respiratory gas flow information to construct patient respiratory status data, and combine normal respiratory status to detect abnormal patient breathing and sensor status. By acquiring information on surgical nodes, the operating status of imaging equipment, and the operating status of instruments, the importance of key surgical nodes is calculated and smoothed to generate key node determination results. Based on the patient's respiratory rate deviation, changes in chest cavity movement, changes in tidal volume, and the status of key surgical nodes, adaptive weights are generated for each respiratory state, and the switching index between spontaneous breathing and mechanical breathing is calculated to determine the current breathing mode and switch to mechanical breathing mode when an abnormality is detected. Based on the determined breathing pattern, a corresponding control command is generated, and the control command is verified to form a data packet for transmission. The data packet is sent through the wireless communication module, and the legality of the received control command is verified. The anesthesia machine is controlled to execute the corresponding breathing mode. At the same time, execution feedback information is generated, and retry and voice prompts are given when execution fails or verification is abnormal. Based on the execution feedback information and the current breathing mode, generate corresponding voice prompts to provide voice feedback during the breathing mode switching process; Establish real-time status data, dynamically sample and update the patient's respiratory status in real time, continuously calculate the switching index between spontaneous breathing and mechanical breathing, and combine abnormal status to determine the dynamic breathing mode. At the same time, perform continuous abnormal safety handling to form a dynamic closed-loop control for breathing switching.

2. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The process involves initializing and self-testing wireless communication parameters, remote control button states, and the voice prompt module. Upon successful remote control initialization, a breathing mode switch is executed. If initialization fails, staff are notified for manual intervention, including: Construct a set of wireless communication parameters that includes multiple communication performance indicators, perform a communication self-test, and determine that the handshake is successful when all indicators meet the preset communication success conditions. Collect the working status of each button on the remote control and confirm that all button functions are normal. The volume, delay, and frequency characteristics of the voice prompt module are self-tested. When each characteristic meets the requirements of human ear recognition and synchronization, the voice self-test is deemed to have passed. The system performs a logical judgment based on three results: successful handshake, all buttons working properly, and voice self-test passed. If the overall result is true, the remote control initialization is successful and the breathing mode switch is executed; otherwise, the initialization fails and staff are notified to intervene manually.

3. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The process involves collecting patient respiratory rate, chest movement amplitude, and respiratory gas flow rate information to construct patient respiratory status data. This data, combined with information on normal respiratory status, is used to detect respiratory abnormalities and sensor status, including: The respiratory airway pressure change waveform is obtained by a pressure sensor, the respiratory cycle is determined according to the time interval between adjacent peaks, and the real-time respiratory rate is calculated from the cycle. The displacement sensor is used to collect chest cavity displacement signals, and the maximum change in chest cavity during each respiratory cycle is extracted as the amplitude of chest cavity movement. Real-time respiratory gas flow is obtained by using a flow sensor, and the tidal volume of a single respiratory cycle is obtained by integrating the flow rate within a single respiratory cycle. The patient's respiratory state vector, composed of the real-time respiratory rate, chest movement amplitude, and tidal volume, is compared with the normal respiratory state vector, which is obtained in advance through statistics and composed of normal respiratory rate, normal chest movement amplitude, and normal tidal volume. If the deviation of any parameter exceeds the preset allowable deviation range, it is determined to be a respiratory abnormality or sensor malfunction.

4. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The process involves acquiring surgical node information, imaging equipment operating status, and instrument operation status to calculate and smooth the importance of key surgical nodes, generating key node determination results, including: Establish a set of surgical nodes and pre-define critical or non-critical status markers for each node; The system acquires the real-time frame rate of intraoperative imaging equipment and automatically generates instrument operation labels based on the key operation time intervals marked by the surgical operator. Based on node markers, image frame rate normalized to standard frame rate, and operation identifiers, a coupled normalization algorithm is used to determine the contribution of node markers, image frame rate, and instrument operation stability, and then generate node marker weight coefficients, image frame rate weight coefficients, and instrument operation state weight coefficients. The importance weight of each node is calculated using the aforementioned weighting coefficients. When the importance weight reaches or exceeds the preset critical node determination threshold, the node is initially determined to be a critical node; The weights of consecutive nodes are smoothed and averaged using a time window of a preset length, and then compared with the threshold again based on the smoothed weights to eliminate the influence of short-term fluctuations and obtain the final key node determination result.

5. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The process involves generating adaptive weights for each respiratory state based on deviations in the patient's respiratory rate, changes in chest movement, changes in tidal volume, and the status of critical surgical points. It also calculates a switching index between spontaneous and mechanical ventilation to determine the current respiratory mode and switches to mechanical ventilation mode when an abnormality is detected. This includes: The respiratory rate potential energy is calculated based on the degree of deviation of the real-time respiratory rate from the normal value. The thoracic motion potential energy is calculated based on the degree of deviation of the chest movement amplitude from the normal value and in combination with the intensity of the surgical node. The surgical node potential energy is calculated based on the intensity of the surgical node and in combination with the degree of deviation of the respiratory rate and the degree of deviation of the chest movement. The sum of the three is used as the total potential field. The respiratory rate weight, chest movement weight, and surgical node weight are calculated based on the respiratory rate potential energy, thoracic cavity motion potential energy, and surgical node potential energy, respectively. The tidal volume potential field is calculated based on the degree of deviation of the tidal volume from the normal value, and the tidal volume weight is determined by combining the respiratory rate potential field, the thoracic motion potential field, and the surgical node potential field. Real-time respiratory rate, chest movement amplitude, and tidal volume are normalized based on their normal values. The switching index between spontaneous breathing and mechanical breathing is calculated using the respiratory rate weight, chest movement weight, tidal volume weight, and key node markers. When the switching index is greater than the set threshold and no abnormal breathing sign appears, the current breathing mode is determined to be spontaneous breathing; otherwise, it is determined to be machine-controlled breathing mode. If signs of abnormal breathing are present, the breathing mode will be forcibly set to mechanical breathing.

6. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The process of generating corresponding control commands based on the determined breathing pattern, and then verifying the control commands to form a data packet for transmission, includes: Generate a corresponding preset binary instruction code based on the determined breathing pattern mapping; If an abnormal flag exists, an emergency stop instruction code is generated. Perform a cyclic redundancy check operation on the instruction code to obtain the check code; The instruction code, checksum, current timestamp, and key node determination flag are combined into a data packet to be sent.

7. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The process involves sending the data packet via a wireless communication module, validating the received control commands, controlling the anesthesia machine to execute the corresponding breathing mode, generating execution feedback information, and providing retry and voice prompts in case of execution failure or verification abnormalities. The data packet is sent using a wireless communication module with preset communication parameters, and the link quality is monitored. When the link quality meets the transmission requirements, the transmission is considered successful. At the receiving end, the cyclic redundancy check code is recalculated on the received instruction code and compared with the received check code. If the two match, the instruction is deemed valid. The breathing mode to be executed is determined according to the valid instruction code, and the corresponding operation is performed by the anesthesia machine; Generate a feedback vector containing the execution success or failure status, timestamp, key node flags, and actual execution mode; When an execution failure or invalid instruction verification occurs, an exception flag is set, and a reminder is issued to the operator via the voice prompt module, while the feedback vector is updated.

8. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The step of generating corresponding voice prompts based on execution feedback information and the current breathing mode to provide voice feedback during the breathing mode switching process includes: Based on the actual breathing pattern and whether the execution was successful or not in the feedback vector, the corresponding voice prompt text code is determined according to the preset mapping relationship; Based on the timestamp in the feedback vector and the preset voice delay parameters, calculate the actual playback time of the voice. Using the set volume and frequency configuration, output corresponding voice prompts at the playback time; Generate a voice playback feedback vector that includes the voice playback success status, playback time, and the encoded text of the played prompt.

9. The control method for a self-service machine-controlled breathing switching device according to claim 1, characterized in that: The establishment of real-time status data, dynamic sampling and real-time updating of the patient's respiratory status, continuous calculation of the spontaneous breathing and mechanical ventilation switching index, and dynamic respiratory mode determination based on abnormal states, along with continuous abnormal safety handling, to form a dynamic closed-loop control for respiratory switching, includes: A real-time state vector is constructed, which includes the patient's real-time respiratory characteristics, chest cavity motion characteristics, tidal volume characteristics, key node markers, and abnormal markers, and is continuously updated according to a preset sampling frequency; The adaptive weights and normalized parameters are used to calculate the current spontaneous breathing and mechanical breathing switching index in real time. Based on whether the switching index is greater than a preset threshold and there is no abnormality indicator, the current breathing mode is dynamically determined to be spontaneous breathing or mechanical breathing. Based on the judgment result, generate and send instruction data packets, receive execution feedback from the anesthesia machine, and record logs for each execution, including timestamps, various status parameters, switching index, breathing mode, and execution status. Perform continuous anomaly safety handling. When the cumulative number of anomaly flags from multiple recent consecutive sampling times reaches the safety threshold, the breathing mode will be forcibly switched to machine-controlled breathing; otherwise, dynamic closed-loop control will be maintained.

Images