Ventilation detection method and device for myasthenia gravis patient

Abstract

The invention relates to a ventilation detection device for a myasthenia gravis patient, and the device comprises a muscle strength monitoring module which is configured to monitor the swallowing muscle of the myasthenia gravis patient and the muscle activity of the near end of the limb, and obtains the muscle strength data of the myasthenia gravis patient; the respiration monitoring module is configured to detect the respiration frequency, the tidal volume of each respiration cycle and the inspiration time of each respiration cycle under the condition that mechanical assisted ventilation is carried out on the myasthenia gravis patient, and the change trend along with time is obtained; the control module is in signal connection with the muscle strength monitoring module and the respiration monitoring module, and is configured to control the respiration monitoring module when the muscle strength data of the serious myasthenia patient is lower than a preset muscle strength threshold value and the change trend of the respiration frequency, the tidal volume of each respiration cycle and the inspiration time along with the time is inconsistent with the change trend of pre-stored corresponding parameters; and determining that man-machine resistance exists in the mechanical assisted ventilation process of the myasthenia gravis patient.

Description

Technical Field

[0001] The present invention relates to the technical field of medical detection equipment, and in particular to a ventilation detection method and device for myasthenia gravis patients. Background Art

[0002] Acquired myasthenia is an autoimmune disease that primarily affects the function of the neuromuscular junction. The cause of acquired myasthenia is that the body's immune system mistakenly attacks the acetylcholine receptors at the neuromuscular junction, causing the nerves to be unable to effectively transmit signals to the muscles. The main symptoms of acquired myasthenia include: muscle weakness, especially of the eyes and facial muscles; drooping eyelids; diplopia (double vision); difficulty swallowing; and difficulty breathing. Existing diagnostic methods for acquired myasthenia include: physical examination; blood tests (antibody tests); neuroelectrophysiological tests; and drug tests (such as the Nestin test). Traditional diagnostic methods rely on clinical evaluation and electrophysiological tests, which are usually not only time-consuming but also somewhat invasive to patients.

[0003] Nguyen, Minh NL, et al. "Tracking eye movements for diagnosis inmyasthenia gravis: a comprehensive review." Journal of Neuro-Ophthalmology 42.4 (2022): 428-441. discusses the potential of quantitative eye and pupil tracking as a non-invasive alternative for diagnosing MG. It highlights how extraocular muscle fatigue can manifest as various abnormalities in eye movements, making eye tracking a valuable tool for early diagnosis, especially in patients presenting with ocular symptoms.

[0004] Qin, Shixin, et al. "Application for measuring eyelid weakness in individuals with myasthenia gravis." 2021 IEEE Global Humanitarian TechnologyConference (GHTC). IEEE, 2021. Discuss a machine learning-based application that can quantitatively track and record eyelid weakness (ptosis), a common symptom of MG. This technology uses eye tracking to help clinicians diagnose MG and monitor the progression of the disease.

[0005] Liang, Timothy, et al. "Analysis of electrooculography signals forthe detection of Myasthenia Gravis." Clinical Neurophysiology 130.11 (2019):2105-2113. evaluated the potential of electrooculography (EOG) signals for the diagnosis of MG. It demonstrated the effectiveness of analyzing EOG signals to distinguish MG patients from controls, further demonstrating the relevance of eye tracking in the diagnosis of MG.

[0006] CN115177291A provides a method for identifying ICU-acquired myasthenia, comprising: obtaining muscle ultrasound data, medical text data and clinical examination data of a patient; based on the muscle ultrasound data, the medical text data and the clinical examination data, identifying the ICU-acquired myasthenia of the patient through a preset multi-interactive memory network. This technical solution can identify the ICU-acquired myasthenia of a patient and its severity without causing muscle trauma.

[0007] The above neurophysiological examination methods for acquired myasthenia have the following defects, such as insufficient sensitivity, low sensitivity of repeated nerve stimulation (RNS) test in patients with ocular myasthenia, which may lead to false negative results, low specificity, strong operation dependence, poor patient tolerance, high equipment requirements, etc. The existing motor function assessment methods for acquired myasthenia also have the defects of strong subjectivity, lack of standardization, limited detection time, difficulty in quantitative detection of disease severity and the impact of the disease on patients' quality of life, and insufficient sensitivity.

[0008] Myasthenia gravis affects the signal transmission between nerves and muscles, causing abnormal fatigue and weakening of skeletal muscles. The disease usually affects the extraocular muscles, facial muscles, throat muscles, and muscles of the limbs. When myasthenia gravis progresses to a more severe stage, it may affect the muscles that control breathing. This condition is called "myasthenic crisis" or "respiratory muscle weakness" and is an acute complication of myasthenia gravis. When the respiratory muscles are affected, the patient may experience difficulty breathing, and in severe cases, even unable to breathe independently, requiring emergency medical intervention, such as the use of a ventilator to assist breathing. Since myasthenia gravis is a disease of the neuromuscular junction, patients may show varying degrees of neuromuscular weakness, which affects their need for ventilator support. Myasthenia gravis patients are more likely to experience human-machine asynchrony when using a ventilator, that is, the patient's spontaneous breathing does not match the assistance provided by the ventilator. This situation may cause discomfort to the patient and even aggravate dyspnea. The prior art lacks a detection device that can judge the patient's spontaneous breathing and whether the patient's spontaneous breathing matches the assistance provided by the ventilator for myasthenia gravis patients. Summary of the invention

[0009] In view of the deficiencies in the prior art, the present application provides a ventilation detection method for myasthenia gravis patients, which is applied to a respiratory assistance system, including: monitoring the muscle activity of the swallowing muscles and the proximal limb muscles of the myasthenia gravis patients; obtaining the muscle strength data of the myasthenia gravis patients based on the monitoring data of the swallowing muscle activity and the proximal limb muscle activity; when the myasthenia gravis patients are subjected to mechanical assisted ventilation, detecting the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time of each respiratory cycle; obtaining the changing trends of the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time over time; when the muscle strength data of the myasthenia gravis patients is lower than a preset muscle strength threshold, and the changing trends of the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time over time are inconsistent with the changing trends of the pre-stored corresponding parameters, it is determined that there is human-machine resistance during the mechanical assisted ventilation of the myasthenia gravis patients.

[0010] According to a preferred embodiment, obtaining muscle strength data of myasthenia gravis patients based on monitoring data of swallowing muscle activity and monitoring data of proximal limb muscle activity includes: collecting muscle strength data of a first time period when the patient performs a first standardized action as reference muscle strength data; collecting muscle strength data of a second time period of a preset time period after the patient performs the first standardized action; and analyzing muscle strength fluctuations based on the muscle strength data of the second time period and the reference muscle strength data.

[0011] According to a preferred embodiment, the first standardized action is swallowing and / or an arm raising action.

[0012] According to a preferred embodiment, after determining that there is human-machine resistance during mechanical assisted ventilation of a myasthenia gravis patient, the method further includes: judging the type of human-machine resistance based on the changing trends of respiratory frequency, tidal volume of each respiratory cycle and inspiratory time over time.

[0013] According to a preferred embodiment, if the increase in respiratory rate is accompanied by fluctuations in the tidal volume waveform, it is determined that there is a failure to effectively trigger the ventilator to deliver air.

[0014] According to a preferred embodiment, if the increase in breathing rate is accompanied by an increase in inspiratory time, it is determined that the ventilator gas delivery is insufficient.

[0015] According to a preferred embodiment, if the tidal volume is lower than a preset tidal volume threshold, the inspiratory time is shorter than a preset inspiratory time threshold and the respiratory rate increases, it is determined that the ventilator has premature exhalation triggering.

[0016] By using the ventilation detection method of the present application, it is possible to determine whether there is human-machine resistance and the type of human-machine resistance according to the characteristics of the condition of myasthenia gravis patients, the muscle strength of myasthenia gravis patients, and the changing trends of tidal volume, respiratory rate, and inspiratory time during mechanically assisted exhalation of myasthenia gravis patients. This method helps the ventilator to promptly detect whether there is human-machine resistance and make timely adjustments when performing mechanical assisted breathing on myasthenia gravis patients.

[0017] The present application also provides a ventilation detection device for myasthenia gravis patients, the device comprising: a muscle strength monitoring module, configured to monitor the muscle activity of the swallowing muscles and the proximal limb muscles of the myasthenia gravis patients, and obtain the muscle strength data of the myasthenia gravis patients according to the monitoring data of the swallowing muscle activities and the proximal limb muscle activities; a respiratory monitoring module, configured to detect the respiratory rate, the tidal volume of each respiratory cycle, and the inspiratory time of each respiratory cycle when the myasthenia gravis patients are subjected to mechanical assisted ventilation, and obtain the respiratory rate, the tidal volume of each respiratory cycle, and the inspiratory time over time. A control module is connected to the muscle strength monitoring module and the respiratory monitoring module signals, and is configured to obtain the muscle strength data of the myasthenia gravis patient from the muscle strength monitoring module, and obtain the respiratory rate, tidal volume of each respiratory cycle, and inspiratory time of the myasthenia gravis patient from the respiratory monitoring module. The changing trend over time, wherein when the muscle strength data of the myasthenia gravis patient is lower than a preset muscle strength threshold, and the changing trends of the respiratory rate, tidal volume of each respiratory cycle, and inspiratory time over time are inconsistent with the changing trends of the corresponding pre-stored parameters, it is determined that there is human-machine resistance in the mechanical assisted ventilation process of the myasthenia gravis patient.

[0018] According to a preferred embodiment, the control module is further configured to determine that the ventilator cannot be effectively triggered to deliver air if an increase in respiratory rate is accompanied by fluctuations in the tidal volume waveform.

[0019] According to a preferred embodiment, the control module is further configured to determine that the ventilator gas delivery is insufficient if an increase in breathing rate is accompanied by an increase in inspiratory time.

[0020] The present application further provides a system that uses wearable devices to monitor patient muscle activity and maintain the patient's respiratory muscles, especially the diaphragm muscle strength. The system can analyze muscle strength data in real time to predict the risk of acquired muscle weakness and perform stimulation of the patient's diaphragm including vibration, compression, electric shock and / or infrared based on the test results.

[0021] The present application provides a breathing assistance system for patients with acquired myasthenia, including the aforementioned ventilation detection device. The system also includes: a chest assistance component, including a first actuator configured for external mechanical contact with the patient's chest and a first control unit connected to the first actuator signal; an abdominal assistance component, including a second actuator configured for external mechanical contact with the patient's abdomen and a second control unit connected to the second actuator signal; wherein the first control unit and the second control unit are configured to cooperate with each other based on the patient's muscle strength data to respectively control the first actuator and the second actuator to provide mechanical stimulation and / or electrical stimulation to the patient's chest and abdomen to assist the patient's breathing.

[0022] According to a preferred embodiment, the system also includes: a first sensor configured to monitor the muscle activity of the swallowing muscles; a second sensor configured to monitor the muscle activity of the proximal limbs; and a data processing unit that receives the detection data of the muscle activity from the first sensor and / or the second sensor and calculates the muscle strength data.

[0023] According to a preferred embodiment, the data processing unit is configured to collect muscle strength data of a first time period when the patient performs a first standardized action through a first sensor and / or a second sensor as reference muscle strength data, and to collect muscle strength data of a second time period of a preset time period after the patient performs the first standardized action through the first sensor and / or the second sensor to analyze muscle strength fluctuations.

[0024] According to a preferred embodiment, the first standardized action includes but is not limited to swallowing action and / or arm raising action, the proximal end of the limb is the patient's arm, the second sensor is configured to detect the movement acceleration of the patient's arm and the equivalent mass of the patient's arm, and the data processing unit calculates the muscle strength data of the user's arm during muscle activity based on the movement acceleration of the patient's arm and the equivalent mass of the patient's arm detected by the second sensor.

[0025] According to a preferred embodiment, the system further comprises a sensing component configured to measure the movement of the patient's respiratory muscles to collect movement information, and determine the movement state of the patient's respiratory muscles based on the collected movement information.

[0026] According to a preferred embodiment, the first control unit and the second control unit are configured to control the first actuator and the second actuator respectively based on the patient's motion state to provide mechanical stimulation and / or electrical stimulation to the patient's chest and abdomen that matches the motion state.

[0027] According to a preferred embodiment, the first actuator includes: a chest strap with adjustable tightness that can surround the chest cavity, a first electrode arranged on the chest strap for stimulating the pectoralis major muscle, a second electrode arranged on the chest strap for stimulating the intercostal muscles, and a first inflatable and deflable mechanical airbag arranged on both sides of the chest strap corresponding to the chest cavity.

[0028] According to a preferred embodiment, the second actuator includes: an abdominal belt with adjustable tightness that can surround the abdominal cavity, a third electrode arranged on the abdominal belt for stimulating the rectus abdominis muscle, a fourth electrode arranged on the abdominal belt for stimulating the external oblique muscle, and an inflatable and deflable second mechanical airbag arranged at a position of the abdominal belt corresponding to the front of the abdomen.

[0029] According to a preferred embodiment, the first actuator and the second actuator are configured to assist the patient in inhaling in the following manner: the first electrode activates the pectoralis major muscle to assist in lifting the thorax, and the second electrode activates the intercostal muscles to help expand the chest cavity; the third electrode and the fourth electrode respectively activate the rectus abdominis and the external oblique muscles to assist in contracting the abdominal muscles and lowering the diaphragm; the first mechanical airbags on both sides of the chest cavity are inflated to assist in expanding the thorax outward; the second mechanical airbag on the front side of the abdomen is contracted to assist in lowering the diaphragm; electrical stimulation and mechanical assistance work together to increase the volume of the chest cavity, reduce the pressure in the chest cavity, and promote the entry of air into the lungs.

[0030] According to a preferred embodiment, the first actuator and the second actuator are configured to assist the patient in exhaling in the following manner: the stimulation intensity of the first electrode and the second electrode is reduced, allowing the chest to fall back naturally; the stimulation intensity of the third electrode and the fourth electrode is increased to assist the contraction of the abdominal muscles, squeeze the abdominal cavity, and push the diaphragm to rise; the first mechanical airbags on both sides of the chest cavity begin to deflate, allowing the chest cavity to retract; the second mechanical airbag on the front side of the abdomen is inflated to assist in squeezing the abdominal cavity; electrical stimulation and mechanical assistance cooperate to increase the pressure in the chest cavity and push the gas out of the lungs.

[0031] The technical effects of the respiratory assistance system for patients with acquired myasthenia in this application include: real-time monitoring and early warning. Through wearable devices, real-time monitoring of swallowing muscle and proximal limb muscle activity can capture small changes in muscle strength in time. By comparing muscle strength data at different times, the system can identify abnormal fluctuations in muscle strength early and provide a basis for early intervention. The system collects muscle strength data when the patient performs standardized movements as a reference to establish personalized baseline data. This method takes into account individual differences and improves the accuracy and pertinence of the assessment, which is specifically manifested in the following aspects. First, dynamic assessment. By comparing muscle strength data at different times, the system can dynamically assess the patient's muscle state and reflect the progression or improvement of the disease. Secondly, respiratory function protection. The system pays special attention to diaphragm function, which is one of the most dangerous complications in patients with acquired myasthenia. Through the respiratory assistance unit, the system can actively maintain diaphragm strength and reduce the risk of respiratory failure. Through multimodal intervention, the respiratory assistance unit provides a variety of stimulation methods, and the most suitable intervention method can be selected according to the specific situation of the patient. Non-invasive monitoring and wearable devices are used for monitoring, which reduces interference with the patient's daily life and improves patient compliance. The continuity and objectivity of the data are guaranteed because the system provides continuous and objective data, overcoming the shortcomings of traditional assessment methods, which are highly subjective and limited in time. Using predictive analysis, by analyzing the fluctuation pattern of muscle strength, the system has the potential to predict the onset or aggravation of acquired myasthenia gravis and provide support for clinical decision-making. At the same time, it expands the possibility of remote monitoring. This system provides a technical basis for remote monitoring, enabling the medical team to grasp the patient's condition in real time and adjust the treatment plan in time. Further, rehabilitation guidance is achieved, and the data collected by the system can be used to guide the patient's rehabilitation training and help develop personalized rehabilitation plans. In general, this system has the potential to significantly improve the management level of acquired myasthenia gravis, improve patient prognosis, and improve the quality of life through real-time, objective, continuous monitoring and timely intervention. It provides an innovative and comprehensive solution for the diagnosis, treatment and rehabilitation of acquired myasthenia gravis.

[0032] According to a preferred embodiment, the first sensor is a flexible sensor attached to the first area of ​​the patient's swallowing muscle skin surface, which obtains a first signal reflecting the change state of the strain of the first area and a second signal characterizing the change state of the curvature of the first area according to the deformation of the first area.

[0033] According to a preferred embodiment, the first sensor includes a strain sensing unit and an optical sensing unit. The strain sensing unit obtains a first signal according to the deformation of the first area. The optical sensing unit obtains a second signal according to the deformation of the first area.

[0034] According to a preferred embodiment, the data processing unit obtains the muscle strength data of the patient's swallowing muscles according to the first signal and the second signal collected by the first sensor.

[0035] According to a preferred embodiment, the data processing unit further includes: a data receiving module for receiving muscle activity detection data from the first sensor and the second sensor; a data analysis module for calculating muscle strength data based on the received muscle activity detection data, analyzing the volatility of muscle strength using a predetermined algorithm model, and outputting a risk assessment result of acquired muscle weakness; and a user interface for displaying the analyzed risk assessment results and suggestions.

[0036] According to a preferred embodiment, the preset time period is adjusted based on the patient's muscle strength data of the first period, historical muscle strength data and / or medical advice.

[0037] According to a preferred embodiment, the data analysis module configures a machine learning algorithm to learn and adapt to the muscle strength fluctuation characteristics of individual patients and update the algorithm model over time. The data analysis module configures an anomaly detection algorithm for detecting and marking abnormal points in muscle strength data. The data analysis module also configures a trend analysis algorithm for identifying long-term trends and short-term fluctuations from muscle strength data over consecutive time periods to assist in assessing the progression of muscle weakness. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] Figure 1 It is a flow chart of a ventilation detection method for myasthenia gravis patients provided by the present invention; Figure 2 It is a schematic diagram of the module structure of the ventilation detection device for myasthenia gravis patients provided by the present invention; Figure 3 is a waveform diagram of an example of “failure to effectively trigger the ventilator to deliver gas” provided in an embodiment of the present invention; Figure 4 is a waveform diagram of an example of "insufficient ventilator gas delivery" provided by an embodiment of the present invention; Figure 5 is a waveform diagram of an example of “early exhalation triggering” provided in an embodiment of the present invention; Figure 6 1 is a schematic diagram of the system structure of a respiratory assistance system for patients with acquired myasthenia provided by the present invention; Figure 7 is an algorithm flow chart for acquired myasthenia gravis provided by the present invention; Figure 8 is a device structure diagram of an embodiment of a respiratory assistance system for patients with acquired myasthenia gravis of the present invention; Fig. 9is a control flow chart of the inhalation phase of an embodiment of the respiratory assistance system for acquired myasthenia patients of the present invention; Fig.10 The present invention is a control flow chart of the exhalation phase of an embodiment of the respiratory assistance system for patients with acquired myasthenia gravis of the present invention.

[0039] Reference numerals list

[0040] 100: chest assistance assembly; 110: first actuator; 120: first control unit; 200: abdominal assistance assembly; 210: second actuator; 220: second control unit; 111: chest belt; 112: first electrode; 113: second electrode; 114: first mechanical airbag; 211: abdominal belt; 212: third electrode; 213: fourth electrode; 214: second mechanical airbag; 130: wearable device; 131: first sensor; 132: communication module; 133: second sensor; 134: EMG sensor; 140: data processing unit; 141: data receiving module; 142: data caching module; 143: data analysis module; 144: processor; 145: memory; 146: storage device; 150: user interface; 151: mobile application; 152: web interface; 153: risk assessment and management plan module; 154: educational resources and support materials; 160: cloud storage server. DETAILED DESCRIPTION

[0041] The following is a detailed description with reference to the accompanying drawings.

[0042] Myasthenia gravis is a chronic autoimmune disease that affects the transmission of signals between nerves and muscles, causing abnormal fatigue and weakening of skeletal muscles. The disease usually affects the extraocular muscles, facial muscles, throat muscles, and muscles in the limbs. In some cases, myasthenia gravis can affect the muscles that control breathing, especially when the disease progresses to a more severe stage. This condition is called "myasthenic crisis" or "respiratory muscle weakness" and is an acute complication of myasthenia gravis. When the respiratory muscles are affected, patients may experience difficulty breathing, and in severe cases, they may not be able to breathe on their own, requiring emergency medical intervention, such as the use of a ventilator to assist breathing.

[0043] There are some special requirements and precautions for patients with myasthenia gravis using ventilators to assist breathing compared to ordinary patients. Since myasthenia gravis is a disease of the neuromuscular junction, patients may show varying degrees of neuromuscular weakness, which will affect their need for ventilator support. Therefore, it is necessary to frequently assess the patient's neuromuscular status, including muscle strength, breathing depth and frequency, so that the ventilator settings can be adjusted in a timely manner. The use of quantitative neuromuscular transmission tests (such as single-fiber electromyography or repetitive nerve stimulation tests) can help better understand the status of neuromuscular function and thus guide treatment decisions.

[0044] The inventor found in the study that during mechanical assisted breathing, myasthenia gravis patients have different muscle strengths, especially breathing-related muscle strengths, due to differences in myasthenia conditions among different patients. The operating parameters of the ventilator are often set according to the doctor's experience. In clinical practice, the ventilator parameter settings often do not match the patient's needs, leading to human-machine resistance, such as failure to effectively trigger the ventilator to deliver air, insufficient ventilator gas delivery, premature exhalation triggering, etc. This situation is more common in myasthenia gravis patients with unstable muscle strength and large differences in muscle strength between individuals.

[0045] The present application provides a ventilation detection method for myasthenia gravis patients, which is applied to a respiratory assistance system, especially a mechanical respiratory assistance system, such as Figure 1 As shown, it includes: monitoring the muscle activity of the swallowing muscles and the proximal limb muscles of the myasthenia gravis patient; obtaining the muscle strength data of the myasthenia gravis patient according to the monitoring data of the swallowing muscle activity and the proximal limb muscle activity; when the myasthenia gravis patient is subjected to mechanical assisted ventilation, detecting the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time of each respiratory cycle; obtaining the changing trends of the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time over time; when the muscle strength data of the myasthenia gravis patient is lower than the preset muscle strength threshold, and the changing trends of the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time over time are inconsistent with the changing trends of the pre-stored corresponding parameters, it is determined that there is human-machine resistance during the mechanical assisted ventilation of the myasthenia gravis patient.

[0046] Tidal Volume (VT): refers to the amount of air that moves in and out of the lungs with each breath, usually expressed in milliliters (mL). The conventional tidal volume for adults is set at 5-8 mL / kg ideal body weight. Respiratory Rate (RR): refers to the number of breaths per minute, used to control minute ventilation (MV), which is the total amount of gas that enters or leaves the lungs in one minute. The common respiratory rate range for adults is 10-20 times / minute. Inspiratory Time (Ti): The length of time the inspiratory phase of each breath lasts. The choice of inspiratory time will affect the I:E ratio and the overall breathing pattern.

[0047] The present application provides a ventilation detection method for myasthenia gravis patients, which aims to evaluate whether human-machine resistance occurs in real time by monitoring the patient's muscle activity and the changing trend of ventilation parameters, and to issue an alarm in time when human-machine resistance occurs. Swallowing muscles are one of the commonly affected parts of myasthenia gravis patients. Monitoring the muscle activity of swallowing muscles can reflect the patient's neuromuscular function status, especially when the respiratory muscles are involved. It is preferred to use a surface electromyography (sEMG) sensor attached to the patient's throat or neck to monitor the electrical activity of swallowing muscles (such as cricopharyngeus, suprahyoid muscles, etc.). These sensors can capture the electrical signals generated when the muscles contract, thereby quantifying muscle strength. For example, the original electromyographic signal can be converted into muscle strength data through signal processing algorithms (such as filtering, denoising, feature extraction, etc.). Common indicators include maximum electromyographic signal strength (RMS), mean frequency (MNF), etc.

[0048] The proximal muscles of the limbs (such as shoulder girdle muscles, hip muscles, etc.) usually have similar neural innervation pathways as the respiratory muscles, so changes in their muscle strength can indirectly reflect the state of the respiratory muscles. It is preferred to use surface electromyography (sEMG) sensors, which are attached to the patient's scapular area, quadriceps, etc. to monitor the electrical activity of the proximal muscles of the limbs. Similar to the swallowing muscles, the electromyographic signals are converted into muscle strength data through signal processing algorithms.

[0049] Preferably, according to clinical experience, the muscle strength of patients with myasthenia gravis usually decreases gradually as the disease progresses. In order to set a reasonable threshold, a baseline measurement can be performed when the patient is admitted to the hospital to record their normal muscle strength level. Subsequently, the threshold is adjusted dynamically according to the changes in the patient's condition. For example, for healthy adults, the maximum myoelectric signal strength (RMS) of the swallowing muscle is usually between 50-100 μV, and the RMS value of the proximal limb muscles is between 100-300 μV. For example, when the patient's muscle strength drops below 50% of the baseline value, it may indicate that neuromuscular function is significantly impaired and there is a higher risk of respiratory failure. For example, if the patient's swallowing muscle RMS value drops below 25 μV, or the proximal limb muscle RMS value drops below 150 μV, an alarm is triggered, indicating that further evaluation of respiratory support is needed.

[0050] The inventors found in their research that there are differences in judging whether human-machine resistance occurs and the possible consequences of human-machine resistance for myasthenia gravis patients with different muscle strength levels. When muscle strength is relatively high, patients can also maintain normal physiological activities through spontaneous breathing combined with ventilator assistance, and weak human-machine resistance generally does not significantly affect the patient's normal breathing. When muscle strength is severely reduced, human-machine resistance may cause serious damage to the patient. According to a specific embodiment, the muscle strength threshold is 30%, 50% or 70% of the baseline value. Medical institutions or detection device manufacturers can set corresponding pre-stored respiratory frequency, tidal volume of each respiratory cycle, and the changing trend of inhalation time according to different muscle strength ranges, and can also set personalized pre-stored respiratory frequency, tidal volume of each respiratory cycle, and the changing trend of inhalation time according to the patient's condition, so as to more accurately judge whether human-machine resistance occurs and evaluate the possible physiological effects on patients.

[0051] Preferably, obtaining muscle strength data of myasthenia gravis patients based on monitoring data of swallowing muscle activity and monitoring data of proximal limb muscle activity includes: collecting muscle strength data of a first time period when the patient performs a first standardized action as reference muscle strength data; collecting muscle strength data of a second time period of a preset time period after the patient performs the first standardized action; and analyzing muscle strength fluctuations based on the muscle strength data of the second time period and the reference muscle strength data.

[0052] Preferably, the first standardized action is a swallowing and / or an arm raising action.

[0053] Preferably, after determining that there is human-machine resistance during mechanical assisted ventilation of myasthenia gravis patients, the method further comprises: judging the type of human-machine resistance according to the changing trends of respiratory frequency, tidal volume of each respiratory cycle and inspiratory time over time.

[0054] Preferably, if Figure 3 As shown, if the increase in respiratory rate is accompanied by fluctuations in the tidal volume waveform, it is determined that the ventilator has not been effectively triggered to deliver gas.

[0055] Preferably, if Figure 4 As shown, if the increase in respiratory rate is accompanied by an increase in inspiratory time, it is determined that the ventilator gas delivery is insufficient, wherein the inspiratory time can be characterized according to the pressure change of the ventilator airway.

[0056] Preferably, if Figure 5 As shown, if the tidal volume is lower than the preset tidal volume threshold, the inspiratory time is shorter than the preset inspiratory time threshold and the respiratory rate increases, it is determined that the ventilator has premature exhalation triggering, wherein the inspiratory time can be characterized according to the pressure change of the ventilator airway.

[0057] Regarding the detection of human-machine resistance when myasthenia gravis patients use ventilators, considering the special pathological characteristics of the neuromuscular junction of such patients, selecting appropriate ventilation parameters is crucial for accurately judging human-machine resistance. Based on the clinical characteristics of myasthenia gravis patients and the need for mechanical ventilation, this application provides a method for monitoring and evaluating human-machine resistance and the type of human-machine resistance using muscle strength, tidal volume (VT), respiratory rate (RR), and inspiratory time (Ti).

[0058] Patients with myasthenia gravis may have weak respiratory muscles, so their tidal volumes may be small or erratic. By monitoring changes in tidal volume, it is possible to identify if the patient is trying to increase their inspiratory effort but is unable to effectively trigger the ventilator to deliver air. Patients with myasthenia gravis may show an increased respiratory rate during an exacerbation. If the respiratory rate is significantly higher than the set value, or there are frequent spontaneous breathing attempts, this may indicate that the patient is struggling to overcome the limitations of the ventilator settings. Inspiratory time reflects how long the inspiratory phase lasts in each breath. Patients with myasthenia gravis may prolong their inspiratory time to obtain enough gas due to weak respiratory muscles. If an abnormally prolonged inspiratory time is observed, or it does not match the set inspiratory to expiratory ratio, it may indicate that there is patient-ventilator resistance.

[0059] Preferably, during mechanical ventilation of the user, data of the above three ventilation parameters are collected in real time and waveforms that change over time are generated. The acquired waveform change trends of tidal volume, respiratory rate and inspiratory time are analyzed, with a focus on whether these parameters show regular abnormal fluctuations or deviations from expected patterns.

[0060] Preferably, the determination method includes: Baseline establishment: First, establish the normal range or baseline waveform for each ventilation parameter based on the patient's baseline condition and current treatment plan. This can be done by collecting data during the initial stabilization period.

[0061] Real-time comparison: Compare the real-time monitored ventilation parameter waveform with the baseline waveform. If the waveform of tidal volume, respiratory rate or inspiratory time is found to be significantly different from the baseline, such as: the tidal volume is too small or too large, accompanied by a significant increase in respiratory rate; the respiratory rate frequently exceeds the set value, accompanied by short and irregular inspiratory attempts; the inspiratory time is abnormally prolonged, resulting in an imbalance in the inspiratory-expiratory ratio; these situations may indicate the occurrence of human-machine resistance.

[0062] Confirming human-machine resistance: When multiple parameters show abnormal trends at the same time, it can be more confidently judged that human-machine resistance exists.

[0063] Preferably, the human-machine resistance type including ineffective inspiratory triggering and premature expiratory triggering is determined by the following method by monitoring and analyzing tidal volume, respiratory rate and inspiratory time.

[0064] Ineffective triggering refers to the patient's attempt to initiate a breath but fails to successfully trigger the ventilator to deliver air, resulting in no response to the respiratory effort. Preferably, ineffective triggering of inspiration is confirmed by tidal volume (VT) waveform analysis and respiratory rate (RR) waveform analysis. Monitor the tidal volume waveform to observe whether there are small tidal volume fluctuations (usually less than 20% of the set tidal volume). These small fluctuations may represent the patient's spontaneous breathing efforts, but due to insufficient trigger sensitivity or improper ventilator settings, they fail to effectively trigger the delivery of air. If the respiratory rate waveform shows frequent short inspiratory attempts, and the tidal volume does not increase significantly after each attempt, it indicates that there may be ineffective inspiratory triggering. Further preferably, the airway pressure waveform can also be additionally checked to find whether there is a negative pressure peak (i.e., the negative pressure generated when the patient inhales), but it is not followed by positive pressure delivery. This indicates that the patient attempted to inhale but failed to trigger the ventilator.

[0065] Premature Cycling Off refers to the situation where the ventilator switches to the expiratory phase too early before the end of the inspiratory phase, resulting in the patient failing to obtain sufficient tidal volume, affecting gas exchange. Monitor the tidal volume waveform to observe whether the tidal volume is significantly lower than the set value. If the tidal volume is often lower than expected, it indicates that there may be premature expiratory triggering. Further preferably, check the airway flow velocity waveform, especially at the end of the inspiratory phase, to observe whether the flow rate drops to the baseline level too early. This indicates that the ventilator switches to the expiratory phase too early, resulting in insufficient inspiratory time. Observe the airway pressure waveform, especially at the end of the inspiratory phase, to see if the pressure drops too early. If the pressure of the inspiratory phase fails to maintain a sufficient time, premature expiratory triggering is indicated.

[0066] Preferably, the patient's respiratory rate is monitored in real time through the built-in flow sensor or pressure sensor of the ventilator. The respiratory rate is usually expressed in breaths per minute (breaths / minute). Under normal circumstances, the respiratory rate of MG patients should be maintained between 10-20 times / minute. If the respiratory rate suddenly increases to more than 25 times / minute and lasts for more than 30 seconds, it may indicate that the patient is struggling to overcome the limitations set by the ventilator. The tidal volume of each respiratory cycle is monitored in real time through the built-in flow sensor of the ventilator. Under normal circumstances, the tidal volume of MG patients should be maintained between 400-600 mL. If the tidal volume drops significantly to below 300 mL and is accompanied by an increase in the respiratory rate, it may indicate that the patient has failed to effectively trigger the ventilator to deliver air, and there is a risk of ineffective inspiratory triggering.

[0067] By using the ventilation detection method of the present application, it is possible to determine whether there is human-machine resistance and the type of human-machine resistance according to the characteristics of the condition of myasthenia gravis patients, the muscle strength of myasthenia gravis patients, and the changing trends of tidal volume, respiratory rate, and inspiratory time during mechanically assisted exhalation of myasthenia gravis patients. This method helps the ventilator to promptly detect whether there is human-machine resistance and make timely adjustments when performing mechanical assisted breathing on myasthenia gravis patients.

[0068] The present application also provides a ventilation detection device for myasthenia gravis patients, such as Figure 2As shown, the device includes: a muscle strength monitoring module, configured to monitor the muscle activity of the swallowing muscles and the proximal limb muscles of the myasthenia gravis patient, and obtain the muscle strength data of the myasthenia gravis patient according to the monitoring data of the swallowing muscle activity and the proximal limb muscle activity; a respiratory monitoring module, configured to detect the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time of each respiratory cycle when the myasthenia gravis patient is subjected to mechanical assisted ventilation, and obtain the changing trend of the respiratory rate, the tidal volume of each respiratory cycle and the inspiratory time over time; a control module, and The muscle strength monitoring module is connected to the respiratory monitoring module signal, and is configured to obtain the muscle strength data of the myasthenia gravis patient from the muscle strength monitoring module, and obtain the respiratory rate, tidal volume of each respiratory cycle, and inspiratory time of the myasthenia gravis patient from the respiratory monitoring module. The change trend over time, when the muscle strength data of the myasthenia gravis patient is lower than the preset muscle strength threshold, and the change trend over time of the respiratory rate, tidal volume of each respiratory cycle, and inspiratory time is inconsistent with the change trend of the corresponding parameters stored in advance, it is determined that there is human-machine resistance during the mechanical assisted ventilation of the myasthenia gravis patient. The inspiratory time of each respiratory cycle is monitored in real time through the built-in flow sensor of the ventilator. Under normal circumstances, the inspiratory time of MG patients should be kept between 0.8-1.2 seconds. If the inspiratory time is significantly extended to more than 1.5 seconds, or shortened to less than 0.6 seconds, it may indicate the risk of delayed exhalation triggering or premature exhalation triggering.

[0069] Preferably, when the patient's muscle strength data (such as the RMS value of the swallowing muscle and the proximal limb muscle) is lower than the preset muscle strength threshold, it indicates that the patient's neuromuscular function is significantly impaired and there may be a risk of respiratory muscle weakness. At the same time, when the change trend of the respiratory rate, tidal volume and inspiratory time over time is inconsistent with the change trend of the corresponding parameters stored in advance, the possibility of human-machine resistance is further confirmed. Specifically, if the patient's muscle strength data is lower than the threshold (such as the swallowing muscle RMS < 25 μV, the proximal limb muscle RMS < 150 μV), and the respiratory rate suddenly increases to more than 25 times / minute, the tidal volume drops below 300 mL, and the inspiratory time is extended to more than 1.5 seconds, the system will determine that there is human-machine resistance and issue an alarm. If the patient's muscle strength data is lower than the threshold, but the change trend of the ventilation parameters is consistent with the pre-stored trend (such as the respiratory rate is stable at 12-16 times / minute, the tidal volume is stable at 450-550 mL, and the inspiratory time is stable at 1.0 seconds), it is considered that there is no obvious human-machine resistance and monitoring continues.

[0070] Preferably, when the patient's muscle strength data is close to the threshold (such as swallowing muscle RMS < 30 μV, proximal limb muscle RMS < 180 μV), but the ventilation parameters have not yet shown obvious abnormalities, medical staff need to pay close attention to the patient's condition. When the patient's muscle strength data is lower than the threshold and the ventilation parameters show abnormal changes, medical staff should immediately check the ventilator settings and take corresponding measures (such as adjusting the trigger sensitivity, increasing PEEP, etc.). When the patient's muscle strength data is far below the threshold (such as swallowing muscle RMS < 20 μV, proximal limb muscle RMS < 120 μV), and the ventilation parameters are seriously abnormal, medical staff should prepare for emergency intubation or adjust the ventilation mode. According to the test results, medical staff can adjust the ventilator settings, such as trigger sensitivity, inspiratory time, ventilation mode, etc., to improve human-machine synchronization. If necessary, combine other clinical intervention measures, such as drug therapy, psychological support, etc., to comprehensively improve the patient's ventilation status.

[0071] Through the above technical solutions, the human-machine resistance phenomenon that may occur in myasthenia gravis patients during mechanical ventilation can be more accurately identified and handled. This solution not only relies on traditional ventilation parameters (such as respiratory rate, tidal volume, and inspiratory time), but also combines the patient's muscle strength data to provide a more comprehensive assessment. Combined with the special pathological characteristics of MG patients and the need for mechanical ventilation, this application provides a safe and effective ventilation management method to ensure that myasthenia gravis patients receive the best respiratory support.

[0072] According to a preferred embodiment, the control module is further configured to determine that the ventilator cannot be effectively triggered to deliver air if an increase in respiratory rate is accompanied by fluctuations in the tidal volume waveform.

[0073] According to a preferred embodiment, the control module is further configured to determine that the ventilator gas delivery is insufficient if an increase in breathing rate is accompanied by an increase in inspiratory time.

[0074] The present application provides a respiratory assistance system for patients with acquired myasthenia. Preferably, the respiratory assistance system includes the aforementioned ventilation detection device. Figure 8 As shown, the system also includes: a chest assist component 100, including a first actuator 110 configured for external mechanical contact with the patient's chest and a first control unit 120 connected to the first actuator 110 by signal; an abdominal assist component 200, including a second actuator 210 configured for external mechanical contact with the patient's abdomen and a second control unit 220 connected to the second actuator 210 by signal; wherein the first control unit 120 and the second control unit 220 are configured to cooperate with each other based on the patient's muscle strength data to respectively control the first actuator 110 and the second actuator 210 to provide mechanical stimulation and / or electrical stimulation to the patient's chest and abdomen to assist the patient's breathing.

[0075] Based on this system, first define the following variables:

[0076] : chest muscle strength as a function of time;

[0077] : Abdominal muscle strength as a function of time;

[0078] : The intensity of stimulation applied by the chest actuator as a function of time;

[0079] : The intensity of stimulation applied by the abdominal actuator as a function of time;

[0080] : Patient's respiratory function index as a function of time.

[0081] The system is described by the following mathematical model:

[0082] The control equation of the chest booster component is:

[0083] ;

[0084] The control equation of the abdominal booster component is:

[0085] ;

[0086] The respiratory function improvement equation is:

[0087] .

[0088] in, , and It is a nonlinear function determined by clinical data and machine learning algorithms.

[0089] According to a specific embodiment, the following linear model is used:

[0090] ;

[0091] ;

[0092] .

[0093] in, , is the control gain coefficient, , They are the maximum muscle strength of the corresponding chest muscles and abdominal muscles of healthy people, , , , is the weight coefficient.

[0094] Specifically, clinical data samples of 10 patients are provided in Table 1 below.

[0095] Table 1

[0096]

[0097] Table 1 lists the clinical data of the 10 patients, recording the patient's serial number, age, gender, muscle weakness, chest muscle strength, abdominal muscle strength, chest stimulation intensity, abdominal stimulation intensity, tidal volume, respiratory rate, and blood oxygen saturation. Among them, chest muscle strength, abdominal muscle strength, chest stimulation intensity, and abdominal stimulation intensity are standardized scores standardized to 0~100. This can be determined by establishing a mapping of chest muscle strength to chest stimulation intensity. This can be determined by establishing a mapping of abdominal muscle strength to abdominal stimulation intensity. It can be determined by establishing a mapping of (chest muscle strength, abdominal muscle strength, chest stimulation intensity, abdominal stimulation intensity) to (tidal volume, respiratory rate, blood oxygen saturation) respectively.

[0098] Preferably, if Figure 8 As shown, the first actuator 110 includes: a chest strap 111 with adjustable tightness and capable of surrounding the chest cavity, a first electrode 112 arranged on the chest strap 111 for stimulating the pectoralis major muscle, a second electrode 113 arranged on the chest strap 111 for stimulating the intercostal muscles, and a first inflatable and deflable mechanical airbag 114 arranged on both sides of the chest strap 111 corresponding to the chest cavity. Specifically, the chest strap 111 is made of elastic fabric such as nylon and spandex blended, and is set as an elastic buckle system, with a chest circumference range of 70 to 130 cm. The first electrode 112 is a surface electrode made of a flexible conductive material such as conductive silica gel, and is set in an elliptical shape, with a long axis of 8 cm and a short axis of 5 cm. It is set on the inner side of the chest strap 111 corresponding to the pectoralis major muscle position, and is connected to the first control unit 120 through a flexible printed circuit. The second electrode 113 is made of a flexible conductive material into a strip electrode, about 15 cm long and about 2 cm wide, and is set on the inner side of the chest strap 111, arranged along the ribs, and connected to the first control unit 120 through a flexible printed circuit. The first mechanical airbag 114 is made of medical grade silicone and is circular with a diameter of about 10 cm. The thickness can be expanded from 0.5 cm to 3 cm. It is embedded in the chest strap structure on both sides of the chest strap 111 and connected to a micro electric pump through a hose for inflation and deflation. A pressure sensor can also be integrated in the first mechanical airbag for real-time pressure monitoring.

[0099] Preferably, if Figure 8As shown, the second actuator 210 includes: an abdominal band 211 with adjustable tightness that can surround the abdominal cavity, a third electrode 212 disposed on the abdominal band 211 for stimulating the rectus abdominis, a fourth electrode 213 disposed on the abdominal band 211 for stimulating the external oblique abdominal muscles, and a second inflatable and deflable mechanical airbag 214 disposed at a position corresponding to the front of the abdomen of the abdominal band 211. The abdominal band 211 is made of breathable elastic fabric, with a width of about 20 cm and a waist circumference range of about 60 to 120 cm. The third electrode 212 is configured as a rectangular surface electrode with a length of 10 cm and a width of 5 cm. It is disposed at a position corresponding to the rectus abdominis on the inner side of the abdominal band 211 and is connected to the second control unit 220 via a flexible printed circuit. The fourth electrode 213 is configured as a fan-shaped surface electrode with an arc length of about 15 cm and a maximum width of about 5 cm. It is disposed at a position corresponding to the external oblique abdominal muscles on the inner side of the abdominal band 211 and is connected to the second control unit 220 via a flexible printed circuit. The second mechanical airbag 214 is made of medical grade silicone and is elliptical with a long axis of 20 cm and a short axis of 15 cm. The thickness can be expanded from 1 cm to 5 cm. It is embedded in the abdominal belt structure at the corresponding position on the front side of the abdomen. It is connected to a micro electric pump through a hose for inflation and deflation. The first mechanical airbag preferably uses the same electric pump as the second mechanical airbag. A pressure sensor can also be integrated in the second mechanical airbag for real-time pressure monitoring.

[0100] The first control unit 120 and the second control unit 220 can adopt a low-power microcontroller (such as ARM Cortex-M4), which has an electrical stimulation module, a programmable current source, an output range of 0~100mA, and an adjustable frequency (1~100Hz); an airbag control module, a PWM-controlled electric pump drive circuit; a communication interface: a Bluetooth low energy (BLE) module for communicating with the main controller; and a sensor interface: which can be used to connect an electromyography (EMG) sensor and a pressure sensor.

[0101] Preferably, if Fig. 9 and Fig.10 As shown, the first actuator 110 and the second actuator 210 are configured to assist the patient's breathing in the following manner:

[0102] Inhalation phase: The first electrode 112 activates the pectoralis major muscle to help lift the thorax, the second electrode 113 activates the intercostal muscles to help expand the thoracic cavity, the third electrode 212 and the fourth electrode 213 lightly activate the rectus abdominis and the external oblique muscles respectively to assist the contraction of the abdominal muscles and lower the diaphragm. The first mechanical airbags 114 on both sides of the thoracic cavity are inflated to assist the thorax to expand outward, and the mechanical airbags on the front side of the abdomen are contracted to assist the diaphragm to descend. Electrical stimulation and mechanical assistance work together to increase the volume of the thoracic cavity, reduce the pressure in the thoracic cavity, and promote the entry of air into the lungs, such as Fig. 9 shown.

[0103] Exhalation phase: The stimulation intensity of the first electrode 112 and the second electrode 113 is reduced, allowing the chest to fall back naturally. The stimulation intensity of the third electrode 212 and the fourth electrode 213 is increased to assist the contraction of the abdominal muscles, squeeze the abdominal cavity, and push the diaphragm up. The first mechanical airbags 114 on both sides of the chest cavity begin to deflate, allowing the chest cavity to retract. The second mechanical airbag 214 on the front side of the abdomen is inflated to assist in squeezing the abdominal cavity. The electrical stimulation and mechanical assistance work together to increase the pressure in the chest cavity and push the gas out of the lungs. Fig.10 shown.

[0104] In this precisely coordinated way, the device is able to mimic the respiratory muscle movement patterns of healthy people, providing comprehensive respiratory assistance to patients with myasthenia gravis. This approach could not only improve patients' respiratory function, but also help maintain and train these muscles, potentially having a positive impact on patients' long-term recovery.

[0105] According to a specific embodiment, the electrode stimulation parameters in the inhalation phase and the exhalation phase are shown in Table 2 and Table 3 below:

[0106] Table 2: Inspiratory phase (duration 1-2 seconds)

[0107]

[0108] Table 3: Exhalation phase (duration 2~3 seconds)

[0109]

[0110] Preferably, the maximum capacity of the first mechanical airbag is 500 ml per side, the inflation rate in the inhalation phase is 250-350 ml / s, and the deflation rate in the exhalation phase is 150-250 ml / s. The maximum capacity of the second mechanical airbag is 1000 ml, the inflation rate in the inhalation phase is 300-400 ml / s, and the deflation rate in the exhalation phase is 200-300 ml / s.

[0111] Preferably, the electrical stimulation intensity of the electrode can be adjusted according to the detected blood oxygen saturation of the patient. According to a specific embodiment,

[0112] ,

[0113] in, is the basic intensity of electrical stimulation, such as listed in Tables 2 and 3, It is the electrode electrical stimulation intensity adjusted according to the patient's blood oxygen saturation. It is the current blood oxygen saturation percentage monitored in real time; is the adjustment factor, with a typical value of 0.02; Indicates the target blood sample saturation value, which is a constant and is usually set to 95~98.

[0114] According to a specific embodiment, the stimulation intensity of the first electrode, the second electrode, the third electrode and the fourth electrode is set according to the patient's detected muscle strength, blood oxygen saturation, respiratory rate and tidal volume.

[0115] ;

[0116] in, Indicates the stimulus intensity, Indicates the standardized muscle strength test value, ranging from 0 to 100. Indicates blood oxygen saturation, ranging from 0 to 100. Indicates respiratory rate, ranging from 12 to 20 times per minute. Indicates tidal volume, usually ranging from 4 to 8 ml / kg body weight. , , , is a parameter set according to the patient's personalized clinical data measurement value. According to a specific embodiment, according to the clinical data measurement value of the test patient, for the first electrode, , , , are set to 0.3, 0.3, 0.2, and 0.2 respectively; for the second electrode, , , , are set to 0.3, 0.25, 0.25, and 0.2 respectively; for the third electrode, , , , are set to 0.2, 0.3, 0.22, and 0.28 respectively; for the fourth electrode, , , , Set to 0.24, 0.28, 0.22, and 0.26 respectively.

[0117] Preferably, the capacity of the first mechanical airbag 114 and the second mechanical airbag 214 can be adjusted according to the patient's vital capacity VC. The maximum capacity of the first mechanical airbag = 0.15×VC; the maximum capacity of the second mechanical airbag = 0.3×VC; wherein VC is the patient's vital capacity (ml).

[0118] Preferably, the upper limit of the electrical stimulation intensity of the electrode is 50 mA. The maximum assisted breathing rate is set to 20 times / minute. An alarm is triggered when it is lower than 90% or no effective breathing is detected for 30 consecutive seconds.

[0119] According to a preferred embodiment, the respiratory assistance system for patients with acquired myasthenia further includes a first monitoring unit for monitoring the thoracic volume of the patient, a second monitoring unit for monitoring the abdominal volume of the patient, a third monitoring unit for collecting the thoracic pressure gradient ΔPt of the patient, and a fourth monitoring unit for collecting the abdominal pressure gradient ΔPa of the patient. The first control unit 120 and the second control unit 220 are configured to adjust the real-time electrical stimulation intensity of the first electrode, the second electrode, the third electrode, and the fourth electrode based on the change of the thoracic volume of the patient collected by the first monitoring unit, the change of the abdominal volume of the patient collected by the second monitoring unit, the change of the thoracic pressure of the patient collected by the third monitoring unit, and the change of the abdominal pressure of the patient collected by the fourth monitoring unit.

[0120] Specifically, the first monitoring unit is a chest circumference measuring resistance belt integrated in the chest belt 111, which can measure the patient's chest circumference changes in real time through resistance changes, thereby calculating the patient's chest cavity volume changes. The second monitoring unit is an abdominal circumference measuring resistance belt integrated in the abdominal belt 211, which measures the patient's abdominal circumference changes in real time through resistance changes, thereby calculating the patient's abdominal cavity volume changes. The third monitoring unit can be an intrathoracic pressure monitor with multiple pressure measuring points to monitor different positions of the chest cavity. The fourth monitoring unit can be an intra-abdominal pressure monitor provided with multiple pressure measuring points to monitor intra-abdominal pressures at different positions.

[0121] Preferably, the first monitoring unit, the second monitoring unit, the third monitoring unit, and the fourth monitoring unit are respectively connected to the first control unit 120 and the second control unit 220 by signal. Specifically, the first monitoring unit, the second monitoring unit, the third monitoring unit, and the fourth monitoring unit are respectively connected to the first control unit 120 and the second control unit 220 by signal through a wireless communication module. The wireless communication module supports at least one communication mode among GPRS, 3G, 4G, 5G, Wi-Fi, ZIGBEE, and LoRa.

[0122] Specifically, the first control unit 120 and the second control unit 220 are configured to calculate the rate of change ΔVt of the patient's thoracic volume based on the patient's thoracic volume collected by the first monitoring unit, calculate the rate of change ΔVa of the patient's abdominal volume based on the patient's abdominal volume collected by the second monitoring unit, and calculate the volume phase difference θv according to the patient's thoracic volume that changes with time t and the patient's abdominal volume that changes with time t.

[0123] According to a specific embodiment, the first control unit 120 and the second control unit 220 are configured to calculate the patient's thoracoabdominal respiratory dynamics synchronization index based on the patient's thoracic volume change rate ΔVt, abdominal volume change rate ΔVa, volume phase difference θv, thoracic pressure gradient ΔPt, and abdominal pressure gradient ΔPa. Among them, the chest and abdomen respiratory dynamics synchronization index Defined as:

[0124] ,

[0125] in, is the volume coordination coefficient, is the pressure balance factor, is the volume-pressure coupling index, is the phase compensation coefficient, is the kinetic efficiency ratio, is the normalized kinetic efficiency ratio.

[0126] Volume coordination coefficient The calculation formula is:

[0127] .

[0128] Pressure balance factor The calculation formula is:

[0129] .

[0130] Volume-pressure coupling index The calculation formula is:

[0131] .

[0132] Phase compensation coefficient The calculation formula is:

[0133] .

[0134] Dynamic efficiency ratio The calculation formula is:

[0135] .

[0136] Normalized kinetic efficiency ratio for:

[0137] .

[0138] It can be further normalized to ensure that its value is between 0 and 100.

[0139] The normalization formula is:

[0140] .

[0141] Among them, the volume coordination coefficient can evaluate the matching degree of chest and abdominal volume changes. It introduces a cosine function correction of the phase difference, and the value range is between 0 and 1. The pressure balance factor reflects the balance of the chest and abdominal pressure gradient, and an exponential function is used to ensure sensitivity. The smaller the pressure difference, the closer the coefficient is to 1. The volume-pressure coupling index is used to evaluate the matching degree of volume changes and pressure changes. The phase compensation coefficient specifically deals with the influence of phase difference, and reaches a maximum value of 1 when the phase difference is 0. The dynamic efficiency ratio can evaluate the efficiency of volume changes and pressure changes, and the Sigmoid function is used for normalization to avoid the influence of outliers. By calculating Respiratory coordination can be quantitatively evaluated, and this parameter can be used to facilitate long-term monitoring and comparison of patients' respiratory stability and recovery status, and can guide respiratory training and treatment.

[0142] According to a specific embodiment, the respiratory assistance system for patients with acquired myasthenia is configured to perform a Adaptive electrical stimulation control scheme for the control of myoelectric signals. First, the basic parameters of electromyographic signal acquisition were set, with a sampling frequency of 2000 Hz, a resolution of 16 bits, a signal-to-noise ratio greater than 60 dB, and a common-mode rejection ratio greater than 100 dB. The first electrode was arranged between the 3rd and 5th intercostal spaces, the second electrode was arranged between the 3rd and 7th intercostal spaces, the third electrode was arranged 5 cm above and below the umbilicus, and the fourth electrode was arranged below the 8th and 9th intercostal margins. The basic parameters of electrode stimulation were pulse width of 200-400 μs, frequency of 20-50 Hz, intensity of 10-40 mA, rise time of 0.5-1 s, duration of 1-2 s, and stimulation interval of 2-4 s.

[0143] Specifically, the first electrode has an initial intensity of 20 mA, a frequency of 35 Hz, and a pulse width of 300 μs. The second electrode has an initial intensity of 15 mA, a frequency of 40 Hz, and a pulse width of 250 μs. The third electrode has an initial intensity of 25 mA, a frequency of 30 Hz, and a pulse width of 350 μs. The fourth electrode has an initial intensity of 22 mA, a frequency of 32 Hz, and a pulse width of 325 μs.

[0144] Further, for example, the training set listed in Table 4 below can be used to train the An intelligent adjustment mechanism for adjusting electrode stimulation parameters. This adjustment mechanism can be customized and adjusted according to the patient's specific physiological state in clinical applications. The following training set is used as the basis An example of feedback regulation. According to a specific implementation, The stimulation intensity and frequency of each of the first electrode, the second electrode, the third electrode and the fourth electrode are adjusted in a manner that is greater than or equal to 80, preferably greater than or equal to 90.

[0145] Table 4

[0146]

[0147] Table 4 lists examples of A training set of technical solutions for adjusting electrode stimulation parameters, including The value and correspondence of The first electrode adjustment value, the second electrode adjustment value, the third electrode adjustment value, and the fourth electrode adjustment value are obtained. By using the training set to train and execute the above feedback adjustment mechanism, it is possible to objectively reflect the patient's respiratory coordination. Parameters are used to evaluate the patient's respiratory status over a long period of time and adjust the electrode stimulation parameters of the respiratory assistance system to achieve better assisted breathing effects.

[0148] Preferably, the system also includes: a first sensor 131, configured to monitor the muscle activity of the swallowing muscles; a second sensor 133, configured to monitor the muscle activity of the proximal limbs; a data processing unit 140, receiving the detection data of the muscle activity from the first sensor 131 and / or the second sensor 133 and calculating the muscle strength data.

[0149] Preferably, the data processing unit 140 is configured to collect muscle strength data of a first time period when the patient performs the first standardized action through the first sensor 131 and / or the second sensor 133 as reference muscle strength data, and to collect muscle strength data of a second time period of a preset time period after the patient performs the first standardized action through the first sensor 131 and / or the second sensor 133 to analyze muscle strength fluctuations.

[0150] Preferably, the first standardized action includes but is not limited to swallowing action and / or arm raising action, the proximal end of the limb is the patient's arm, the second sensor 133 is configured to detect the movement acceleration of the patient's arm and the equivalent mass of the patient's arm, and the data processing unit 140 calculates the muscle strength data of the user's arm during muscle activity based on the movement acceleration of the patient's arm and the equivalent mass of the patient's arm detected by the second sensor 133.

[0151] Preferably, the system further comprises a sensing component configured to measure the action of the patient's respiratory muscles to collect action information, and determine the action state of the patient's respiratory muscles based on the collected action information.

[0152] Preferably, the first control unit 120 and the second control unit 220 are configured to respectively control the first actuator 110 and the second actuator 210 based on the motion state of the patient to provide mechanical stimulation and / or electrical stimulation to the chest and abdomen of the patient that matches the motion state.

[0153] Preferably, the electrode is configured to adjust the intensity of the electrical stimulation in turn by the detected muscle strength fluctuation rate and the real-time detected blood oxygen saturation. According to another specific embodiment, the system is configured to give priority to a control method for adjusting the electrode electrical stimulation intensity based on the real-time detected blood oxygen saturation when the detected blood oxygen saturation is lower than 95. When the detected blood sample saturation is higher than 95, a control method for adjusting the electrode electrical stimulation intensity based on the detected muscle strength fluctuation rate is given priority. This control strategy cleverly balances the patient's immediate safety needs (by monitoring blood oxygen saturation) and long-term treatment effects (by monitoring muscle strength). This balance is particularly important for the management of patients with chronic diseases. By adjusting treatment parameters in real time, the system can provide tailored treatment according to the real-time condition of each patient. The system can not only respond to acute situations (such as a sudden drop in blood oxygen), but also prevent potential problems (such as excessive fatigue) by monitoring muscle status. This preventive approach may greatly reduce the risk of complications.

[0154] Long-term use of this system will accumulate a large amount of valuable data, which can be used to further optimize treatment strategies and may even provide new insights into the study of related diseases. By optimizing the intensity and frequency of stimulation, the system may significantly improve the patient's comfort and quality of daily life, which is critical for patients who use respiratory assistance devices for a long time. This control strategy demonstrates how to integrate multiple physiological parameters into an intelligent decision-making system, providing inspiration for the design of other medical devices.

[0155] Preferably, the first actuator 110 is configured to apply vibrations along the intercostal space of the patient to enhance the inhalation process. Preferably, the first actuator 110 is configured to apply pressure to the pectoralis major and pectoralis minor muscles to enhance the exhalation process. Preferably, the second actuator 210 is configured to apply pressure or vibration to the external and internal oblique abdominal muscles on the side of the patient's abdomen. Preferably, the second actuator 210 is configured to apply circumferential pressure or vibration to the transverse abdominal muscle at the patient's waist.

[0156] Preferably, for the intercostal muscles, a small vibrating device can be used to stimulate along the intercostal spaces. This helps to enhance the inhalation process. For the pectoralis major and minor, a larger vibration pad or pressure pad can be used to cover the entire front of the chest. This helps to enhance the exhalation process. For the rectus abdominis, a belt vibrator or pressure belt can be used to stimulate from below the sternum to above the pubic bone. For the external and internal obliques, a diagonal stimulation belt can be used on the side of the abdomen. Stimulation of the transverse abdominis can be achieved by using a wrap-around stimulation belt around the waist.

[0157] Preferably, design a pattern that alternates the stimulation of the chest and abdomen to mimic the rhythm of natural breathing. For example, stimulate the chest muscles for 1 to 2 seconds (mimicking inhalation) and then stimulate the abdominal muscles for 2 to 3 seconds (mimicking exhalation). Use low intensity and low frequency at the beginning, and gradually increase to a level that the patient can accept comfortably. The frequency can start from 12 to 20 times per minute to mimic the normal breathing rate.

[0158] The system can analyze muscle strength data in real time to predict the risk of acquired muscle weakness and stimulate the patient's diaphragm, including vibration and squeezing, based on the test results. Figure 6 As shown, the present application provides a life support system for acquired muscle weakness. The system includes: a first sensor 131, configured to monitor the muscle activity of the swallowing muscle; a second sensor 133, configured to monitor the muscle activity of the proximal limb; a data processing unit 140, receiving the detection data of the muscle activity from the first sensor 131 and / or the second sensor 133 and calculating the muscle strength data; a breathing assistance unit, configured to assist the patient's breathing by vibration, compression, electric shock and / or infrared. The data processing unit 140 is configured to collect the muscle strength data of the first time period when the patient performs the first standardized action through the first sensor 131 and / or the second sensor 133 as the reference muscle strength data, collect the muscle strength data of the second time period of the preset time period after the patient performs the first standardized action through the first sensor 131 and / or the second sensor 133 to analyze the muscle strength fluctuation, and the breathing assistance unit stimulates the patient's diaphragm according to the muscle strength fluctuation to maintain the patient's diaphragm strength.

[0159] Preferably, the first sensor 131 and the second sensor 133 are configured in the wearable device. The data processing unit 140 is in data communication with the first sensor 131 and the second sensor 133. The data processing unit 140 divides and analyzes the data of the first sensor 131 and the second sensor 133 according to time. Preferably, the system further includes a user interface 150 for displaying muscle weakness and risk assessment results. Preferably, the wearable device further includes: at least one electromyography (EMG) sensor 134 for monitoring and analyzing comprehensive muscle activity. The sensor sends data to the data processing unit 140 via a wireless protocol. Preferably, the data processing unit 140 further includes: a set of advanced data processing algorithms for extracting and analyzing key features in sensor data; a set of machine learning and deep learning algorithms for refining prediction models and enhancing system performance. Preferably, the user interface 150 includes: a mobile application 151 or a web interface 152 that allows patients to access and interact; a personalized risk assessment and management plan module 153, customized according to individual patient needs and conditions; educational resources and support materials 154, designed to provide knowledge and self-management strategies to patients.

[0160] The interaction between the data processing unit 140 and the wearable device 130 is optimized as follows: data transmission protocol and communication protocol to ensure real-time transmission and real-time processing of muscle strength data; secure data encryption and authentication mechanism to protect patient privacy and data integrity; data synchronization and integration function with the cloud storage server 160 for data backup and remote access.

[0161] Preferably, the system further includes an enhanced connection to a cloud storage server 160, including: strong data security measures such as access control, data encryption, and regular security audits; scalable storage capacity to accommodate the growth of patient data volume; and data analysis tools for extracting valuable insights from accumulated patient data to promote system improvement and myasthenia gravis research.

[0162] The specific algorithm for evaluating muscle weakness involved in this application is as follows Figure 7 The specific instructions are as follows:

[0163] Symbol Definition

[0164] : No. Time period

[0165] : No. Muscle strength data for each period

[0166] : Standardized muscle strength data set for the first period

[0167] : No. Period and Middle Muscle strength data associated with each data

[0168] Judgment process

[0169] 1. Standardize the first period data

[0170] The muscle strength data of the first period Perform standardization to obtain a standardized set :

[0171] ,

[0172] in, yes The mean of is the standard deviation.

[0173] 2. Calculate muscle strength fluctuations

[0174] For each subsequent period (i > 1), calculate its muscle strength data and The difference in the corresponding data:

[0175] .

[0176] 3. Determine acquired muscle weakness

[0177] if Exceeding the preset threshold , it is judged that acquired muscle weakness exists during this period.

[0178] The choice of threshold can be adjusted based on clinical experience or the patient's personal historical data. At the same time, data noise, individual differences of patients, and joint analysis of multi-muscle group data should be considered to optimize the judgment accuracy.

[0179] Specifically, the clinical test data of several patients are listed in Table 5 below.

[0180] Table 5

[0181]

[0182] Table 5 lists the clinical test data of the two patients, including the corresponding swallowing muscle EMG data, arm acceleration data, measured arm equivalent mass, calculated muscle strength data, respiratory rate and tidal volume data at the corresponding time points when the two patients with serial numbers 1 and 2 performed swallowing or hand-raising movements at two time points, T0 and T1, respectively. Swallowing muscle strength is approximately linearly correlated with EMG signal strength and can be calculated using the following formula:

[0183] ;

[0184] in, is the calculated swallowing muscle strength (unit: N), is the proportionality factor (unit: N / μV), which needs to be determined through calibration. is the measured EMG signal intensity (unit: μV).

[0185] According to a specific embodiment, swallowing muscle strength is measured using a surface electromyography sensor, i.e., EMG sensor 134, to record the amplitude and frequency of the electromyographic signal during swallowing. The measurement parameters of arm muscle strength measurement are arm movement acceleration a and arm equivalent mass m. Muscle strength F=m×a. Among them, F is muscle strength (Newton, N), m is arm equivalent mass (kilograms, kg), and a is arm acceleration (meters / second², m / s²).

[0186] The test process is as follows: First period (reference data collection): perform standardized actions (such as swallowing or raising the arm) and record muscle strength data as a reference. Second period (fluctuation analysis): perform standardized actions again after a preset period of time, record muscle strength data, compare with reference data, and analyze strength fluctuations. Strength fluctuations are characterized by the calculated strength fluctuation rate R. R=(F 第二时段 -F 第一时段 ) / F 第一时段 × 100%. R represents the power fluctuation rate, F 第二时段 Indicates the muscle strength tested in the second period, F 第一时段 Indicates the muscle strength detected in the first period.

[0187] The relationship between the intensity of electrical stimulation and the fluctuation rate of muscle strength is: 调整后 = I 基础 ×(1-k×R); where I 基础 is the basic intensity of electrical stimulation, such as listed in Table 2 and Table 3, I 调整后 It is the electrode electrical stimulation intensity adjusted according to the patient's power fluctuation rate. R represents the power fluctuation rate, and k is the adjustment coefficient, with a value range of 0.1~0.3.

[0188] Preferably, the second actuator 210 is an abdominal pressure device, which receives the muscle strength fluctuation information of the patient's swallowing muscles and proximal upper limbs from the data processing unit 140, and presses the patient's abdomen in a bionic breathing exercise manner to maintain the patient's diaphragm and lung activity. Preferably, the pressing mode and strength are also set in association with the patient's intra-abdominal pressure. The patient's intra-abdominal pressure can be provided by an intra-abdominal pressure detection device.

[0189] For example, in the muscle strength fluctuation detection, the system detected that the strength of the patient's swallowing muscles decreased by 15% in 10 minutes (from the initial 100N to 85N). At the same time, the proximal muscle strength of the upper limbs decreased by 20% in 15 minutes (from the initial 150N to 120N). Based on these data, the abdominal pressure device was activated and began to press the patient's abdomen in a bionic breathing movement. The initial compression frequency was set to 12 times per minute to mimic the normal breathing frequency. The initial compression depth was set to 2cm to simulate mild breathing movements. The intra-abdominal pressure detection device measured the patient's basic intra-abdominal pressure to be 8 mmHg. The system sets the intra-abdominal pressure fluctuation range during normal breathing to 6~10 mmHg. If the intra-abdominal pressure is detected to drop to 5 mmHg, the system will increase the compression force to a depth of 3cm and the frequency to 14 times per minute. If the intra-abdominal pressure rises to 12 mmHg, the system will reduce the compression force to a depth of 1.5cm and the frequency to 10 times per minute. The system evaluates muscle strength and intra-abdominal pressure every 5 minutes and dynamically adjusts compression parameters. If muscle strength recovers to more than 95% of the initial value within 15 minutes (swallowing muscle strength recovers to 95N, and proximal upper limb muscle strength recovers to 142.5N), the system will gradually reduce the compression intensity and frequency. The system sets the maximum compression depth to no more than 4cm to prevent pressure on internal organs. The maximum compression frequency is limited to 20 times per minute to avoid hyperventilation. Based on the patient's height, weight, and vital capacity, the system can personalize the initial parameters. For example, for a patient with a height of 170cm and a weight of 65kg, the initial compression depth may be set to 2.2cm. Through this precise, personalized dynamic adjustment, the respiratory assistance system can more effectively maintain the patient's diaphragmatic function and respiratory capacity while reducing the risk of complications.

[0190] According to a specific embodiment, the data processing unit 140 is configured to control the actuator to perform variable frequency vibration with a non-constant frequency in the operating frequency range of 5 Hz to 1000 Hz during a given time interval to assist in maintaining the patient's diaphragm muscle strength. For example, the actuator specifications are weight: 50 grams, size: 5cm × 3cm × 1cm, maximum output power: 2 watts. Vibration frequency range: minimum frequency: 5Hz, maximum frequency: 1000Hz, frequency adjustment accuracy: 1Hz. Vibration modes include: low frequency band (5Hz~50Hz): starting frequency: 10Hz, lasting 5 seconds, gradually increasing to 30Hz, lasting 10 seconds, and decreasing to 20Hz, lasting 5 seconds. Medium frequency band (51Hz~500Hz): starting from 100Hz, increase 50Hz every 2 seconds until reaching 300Hz, maintain at 300Hz for 5 seconds, and then decrease 100Hz every 3 seconds until returning to 100Hz. High frequency band (501Hz~1000Hz): Start from 600Hz, quickly increase to 900Hz (within 1 second), maintain at 900Hz for 2 seconds, slowly decrease to 700Hz (within 3 seconds), and repeat this cycle 3 times. Time interval setting: Total treatment time: 15 minutes, low frequency band: 5 minutes, medium frequency band: 7 minutes, high frequency band: 3 minutes. Preferably, the system evaluates the patient's diaphragm electromyography (EMG) signal every 30 seconds. If the EMG signal intensity decreases by more than 10%, the system increases the vibration intensity and duration. For example: If a decrease in the EMG signal is detected in the medium frequency band, the system may increase the maintenance time of 300Hz from 5 seconds to 8 seconds. Based on the patient's weight and muscle condition, the system can adjust the vibration intensity. For example, for a patient weighing 60kg, the initial vibration intensity may be set to 1.5 watts, and for a patient weighing 80kg, the initial vibration intensity may be set to 1.8 watts. Continuous high-frequency vibration (>800Hz) does not exceed 30 seconds. After every 5 minutes of treatment, the system is forced to pause for 30 seconds to prevent muscle fatigue. If the patient reports discomfort (via button or voice command), the system immediately reduces the frequency by 50%. If the discomfort persists, the system gradually reduces the frequency. Through this precise, dynamic vibration stimulation, the Breathing Assist System is able to effectively maintain and stimulate the patient's diaphragm muscle strength while providing a personalized and safe treatment plan. This approach may help prevent respiratory impairment in patients with acquired myasthenia gravis and improve their quality of life.

[0191] Preferably, the second actuator 210 also comprises a belt of at least two vibration modules, which are externally applied to the abdominal area of ​​the user to stimulate the diaphragm, thereby enhancing lung function.

[0192] Preferably, the first sensor 131 is a flexible sensor attached to the first area of ​​the patient's swallowing muscle skin surface, which obtains a first signal reflecting the strain change state of the first area and a second signal representing the curvature change state of the first area according to the deformation of the first area.

[0193] Preferably, the first sensor 131 includes a strain sensing unit and an optical sensing unit. The strain sensing unit obtains a first signal according to the deformation of the first area. The optical sensing unit obtains a second signal according to the deformation of the first area.

[0194] Preferably, the data processing unit 140 acquires muscle strength data of the patient's swallowing muscles according to the first signal and the second signal collected by the first sensor 131 .

[0195] Preferably, a flexible material (such as polydimethylsiloxane PDMS) is used to make the sensor substrate to ensure that it fits the skin and does not affect the swallowing action. The strain sensing unit is made of nanomaterials such as graphene or carbon nanotubes, which has high sensitivity and good flexibility. The optical sensing unit can use fiber grating (FBG) technology to accurately measure small changes in curvature. The sensor surface is coated with biocompatible materials to reduce irritation to the skin. The sensor can fit the surface of the swallowing muscle tightly and accurately capture muscle movement. The flexible design will not affect the patient's normal swallowing, and improves the comfort of long-term monitoring. The strain sensing unit measures muscle deformation through resistance changes and converts it into an electrical signal output. The optical sensing unit uses the phase change of the light wave in the optical fiber to measure the curvature and converts it into an optical signal output. A high-precision analog-to-digital converter (ADC) is used to convert the analog signal into a digital signal. A low-noise amplifier circuit is designed to improve the signal quality. In this way, strain and curvature information can be obtained simultaneously to fully reflect the movement state of the swallowing muscle. A high sampling rate (such as 1000Hz) can capture fast swallowing movements. Use digital filtering algorithms (such as Butterworth filters) to remove environmental noise and baseline drift. Apply machine learning algorithms (such as support vector machines (SVMs) or deep neural networks) to extract features from the raw signal. Convert strain and curvature data into muscle strength data to establish a mathematical model to achieve real-time data processing and provide immediate feedback. In this way, the muscle strength of the swallowing muscles can be accurately estimated, providing an objective basis for clinical evaluation.

[0196] Preferably, the data processing unit 140 further includes: a data receiving module 141, used to receive muscle activity detection data from the first sensor 131 and the second sensor 133; a data analysis module 143, used to calculate muscle strength data based on the received muscle activity detection data, use a predetermined algorithm model to analyze the volatility of muscle strength, and output a risk assessment result of acquired muscle weakness; a user interface 150, used to display the analyzed risk assessment results and suggestions.

[0197] Preferably, the data processing unit 140 is connected to the wearable device, and is used to receive muscle strength data from the first sensor 131 and / or the second sensor 133, and analyze the data to predict the risk of acquired muscle weakness; wherein the muscle strength data is divided according to multiple different time periods, specifically: muscle strength data of the first time period, that is, reference muscle strength data collected when the patient performs the first standardized action; muscle strength data of the second time period, that is, muscle strength data collected after a preset time period after the patient performs the first standardized action, and used to predict muscle strength fluctuations.

[0198] In this way, the wearable device integrates the first sensor 131 and the second sensor 133 for collecting muscle strength data. The data processing unit 140 is wirelessly connected to the wearable device to receive and analyze the muscle strength data. In the first period: the patient performs a first standardized action (such as a grip test or a swallowing test) and collects reference muscle strength data. In the second period: the patient performs the same action again in a preset time period (such as 30 minutes) after performing the first standardized action, and collects muscle strength data for prediction. The data processing unit 140 first performs data preprocessing, filters, denoises, and other processing on the original sensor data, and then performs feature extraction, extracts key features from the processed data, such as peak strength, duration, decay rate, etc., and then performs data comparison, compares the muscle strength data of the first period and the second period, calculates the percentage of strength change, and finally performs risk assessment. Based on the data comparison results, a machine learning algorithm (such as a support vector machine or a random forest) is used to predict the risk of acquired myasthenia gravis and output a stimulation mode according to a preset model to start the corresponding respiratory assistance device. The stimulation mode includes the selection of compression, vibration, electric shock or infrared mode and the specific setting parameters of the corresponding mode. In this technical solution, by comparing the changes in muscle strength in a short period of time, the potential risk of acquired myasthenia can be discovered early, which can win valuable time for clinical intervention. The use of wearable device 130 for detection reduces the discomfort of traditional examination methods and improves patient compliance. Through this non-invasive wearable device detection, 24-hour uninterrupted monitoring can be achieved to fully grasp the changes in the patient's muscle strength. Taking into account individual differences, the system can perform customized risk assessment based on the patient's personal baseline data. The data can be transmitted to medical institutions in real time, which is convenient for doctors to remotely monitor the patient's condition and adjust the treatment plan in time. This technical solution is suitable for data accumulation and analysis, and the data collected over a long period of time is helpful for in-depth research on the pathogenesis and influencing factors of acquired myasthenia. Through standardized data collection and analysis processes, subjective judgment errors are reduced, and the objectivity and accuracy of diagnosis are improved. Through early warning and timely intervention, the occurrence of serious complications may be reduced, thereby reducing overall medical costs. This acquired myasthenia risk prediction system based on wearable device 130 and data analysis and respiratory / life support system can significantly improve the level of prevention, diagnosis and rehabilitation treatment of acquired myasthenia, and provide better medical protection for patients.

[0199] Preferably, the wearable device further includes a communication module 132 for wirelessly transmitting the muscle strength data collected by the first sensor 131 and the second sensor 133 to the data processing unit 140 .

[0200] Preferably, the preset time period is adjusted based on the patient's muscle strength data of the first period, historical muscle strength data and / or medical advice.

[0201] Preferably, the data processing unit 140 further communicates with a cloud storage server 160 to transmit the analyzed muscle strength fluctuation data and risk assessment results to the cloud storage server 160 for storage; receive historical data from the cloud storage server 160 to assist in the analysis and judgment of current data and improve the accuracy of prediction; allow authorized medical personnel to access the patient's historical and current muscle strength data for remote diagnosis and treatment planning.

[0202] Preferably, the communication module 132 supports at least one wireless communication technology, including but not limited to Bluetooth, Wi-Fi, NFC or cellular network technology.

[0203] Preferably, the user interface 150 is a touch screen for providing interactive operations, guiding the patient to complete standardized movements through the user interface 150, and displaying muscle strength data and evaluation results in real time.

[0204] Preferably, the data receiving module 141 and the data analysis module 143 are embedded software modules, running on the hardware platform of the data processing unit 140, wherein the hardware platform of the data processing unit 140 includes but is not limited to a processor 144, a memory 145 and a storage device 146. Preferably, the data processing unit 140 is further configured with a data cache module 142 for temporarily storing real-time muscle strength data received from the wearable device to ensure that the data is not lost when the wireless network is unstable or interrupted. Preferably, the data analysis module 143 is configured with a machine learning algorithm to learn and adapt to the muscle strength fluctuation characteristics of individual patients and update the algorithm model over time. The data analysis module 143 is configured with an anomaly detection algorithm for detecting and marking abnormal points in muscle strength data. The data analysis module 143 is also configured with a trend analysis algorithm for identifying long-term trends and short-term fluctuations from muscle strength data of continuous time periods to assist in assessing the progression of muscle weakness.

[0205] It should be noted that the above-mentioned specific embodiments are exemplary, and those skilled in the art can come up with various solutions inspired by the disclosure of the present invention, and these solutions also belong to the disclosure scope of the present invention and fall within the protection scope of the present invention. Those skilled in the art should understand that the present invention specification and its drawings are illustrative and do not constitute a limitation on the claims. The scope of protection of the present invention is defined by the claims and their equivalents. The present invention specification contains multiple inventive concepts, such as "preferably", "according to a preferred embodiment" or "optionally", which means that the corresponding paragraph discloses an independent concept, and the applicant reserves the right to file a divisional application based on each inventive concept.

Claims

1. A ventilation detection device for myasthenia gravis patients, characterized in that: The device comprises: a muscle strength monitoring module configured to monitor the muscle activity of the swallowing muscles and the muscle activity of the proximal limbs of the myasthenia gravis patient, and obtain the muscle strength data of the myasthenia gravis patient according to the monitoring data of the muscle activity of the swallowing muscles and the monitoring data of the muscle activity of the proximal limbs; a respiratory monitoring module configured to detect the respiratory rate, the tidal volume of each respiratory cycle, and the inspiratory time of each respiratory cycle when the myasthenia gravis patient is subjected to mechanical assisted ventilation, and obtain the changing trends of the respiratory rate, the tidal volume of each respiratory cycle, and the inspiratory time over time; The control module is connected to the muscle strength monitoring module and the respiratory monitoring module by signal, and is configured to obtain the muscle strength data of the myasthenia gravis patient from the muscle strength monitoring module, and obtain the respiratory rate, tidal volume of each respiratory cycle, and the change trend of the inspiratory time over time of the myasthenia gravis patient from the respiratory monitoring module. Among them, when the muscle strength data of the myasthenia gravis patient is lower than the preset muscle strength threshold, and the changing trends of the respiratory rate, tidal volume of each respiratory cycle and inspiratory time over time are inconsistent with the changing trends of the pre-stored corresponding parameters, it is determined that there is human-machine resistance during the mechanical assisted ventilation process of the myasthenia gravis patient.

2. The ventilation detection device for myasthenia gravis patients according to claim 1, characterized in that: The control module is further configured to determine that the ventilator cannot be effectively triggered to deliver air if an increase in respiratory rate is accompanied by fluctuations in the tidal volume waveform.

3. The ventilation detection device for myasthenia gravis patients according to claim 1, characterized in that: The control module is further configured to determine that ventilator gas delivery is insufficient if an increase in respiratory rate is accompanied by an increase in inspiratory time.

4. A ventilation detection method for myasthenia gravis patients, applied to a respiratory assistance system, characterized in that: The method comprises: Monitor the swallowing muscle activity and proximal limb muscle activity in patients with myasthenia gravis; Obtaining muscle strength data of myasthenia gravis patients based on monitoring data of swallowing muscle activity and monitoring data of proximal limb muscle activity; In patients with myasthenia gravis receiving mechanical ventilation, respiratory rate, tidal volume per respiratory cycle, and inspiratory time per respiratory cycle were measured; Obtain the time trend of respiratory rate, tidal volume per respiratory cycle and inspiratory time; When the muscle strength data of a patient with myasthenia gravis is lower than a preset muscle strength threshold, and the changing trends of the respiratory rate, tidal volume of each respiratory cycle, and inspiratory time over time are inconsistent with the changing trends of the corresponding pre-stored parameters, it is determined that human-machine resistance exists during the mechanical assisted ventilation of the patient with myasthenia gravis.

5. The ventilation detection method for myasthenia gravis patients according to claim 4, characterized in that: The muscle strength data of myasthenia gravis patients obtained based on the monitoring data of swallowing muscle activity and proximal limb muscle activity include: collecting muscle strength data of a first period of time when the patient performs a first standardized action as reference muscle strength data; Collecting muscle strength data of a second period of a preset time period after the patient performs the first standardized action; The muscle strength fluctuation is analyzed based on the muscle strength data of the second period and the reference muscle strength data.

6. The ventilation detection method for myasthenia gravis patients according to claim 5, characterized in that: The first standardized maneuver is swallowing and / or arm raising.

7. The ventilation detection method for myasthenia gravis patients according to claim 4, characterized in that: After determining that there is resistance during mechanical assisted ventilation in the myasthenia gravis patient, the method further includes: The type of human-machine resistance was determined based on the changing trends of respiratory rate, tidal volume of each respiratory cycle, and inspiratory time over time.

8. The ventilation detection method for myasthenia gravis patients according to claim 7, characterized in that: If the increase in respiratory rate is accompanied by fluctuations in the tidal volume waveform, it is judged that the ventilator has not been effectively triggered to deliver gas.

9. The ventilation detection method for myasthenia gravis patients according to claim 7, characterized in that: If the increase in respiratory rate is accompanied by an increase in inspiratory time, the ventilator is judged to be insufficient in gas delivery.

10. The ventilation detection method for myasthenia gravis patients according to claim 7, characterized in that: If the tidal volume is lower than the preset tidal volume threshold, the inspiratory time is shorter than the preset inspiratory time threshold and the respiratory rate increases, it is determined that the ventilator has premature expiratory triggering.

Images