Study Of Key Algorithms For Monitoring And Curing Sleep Apnea Syndrome

Sleep apnea syndrome is a chronic disease, seriously affecting people’s health. Sleep apnea syndrome is divided into OSAS, CSAS and and MSAS. Clinically, OSAS is most common. In this paper, the treatment and monitoring of OSAS is a major.Clinical manifestations of obstructive sleep apnea syndrome (OSAS) have nocturnal snoring, sleep apnea and daytime sleepiness associated. Because of repeated episodes of apnea caused by nocturnal hypoxia and hypercapnia can lead to hypertension, coronary heart disease, diabetes, cerebrovascular diseases and other complications, accidents, and even sudden death at night. So OSAS is a potentially fatal sleep breathing disorder.Direct pathogenesis of OSAS is that airway is narrow and blocking, but its incidence is not simply because of airway obstruction, actually it is the upper airway collapse, accompanied by breathing regulator of central nervous system disorders. Many causes narrowing and obstruction of the upper airway, including the nasal septum, hypertrophy of the tonsils, soft palate too long, narrow mandibular arch, mandibular retrusion deformities, temporomandibular joint, both sides appeared in a few cases a small jaw ankylosis secondary deformity, macroglossia, after hyoid shift and so on. In addition, obesity, upper airway mucous tissue edema and oropharynx or hypopharynx cancer, also can cause OSAS. The etiology and pathogenesis of OSAS still need further study.OSAS treatment mainly behavioral therapy, noninvasive ventilation therapy, surgery and drug treatment and other methods. Non-invasive ventilation therapy use non-invasive ventilator, which the way of treatment is continuous positive airway pressure (CPAP), bilevel positive airway pressure (BiPAP) and automatic positive airway pressure (APAP). Ventilator therapy growing in popularity is due to not need surgery and medication and It ensures that the effect of non-invasive treatment. Ventilator for the treatment of obstructive sleep apnea syndrome (OSAS) is safe and effective. With advances in technology, more and more types of ventilator have emerged. Requirements for ventilator is not limited therapeutic effect, but aiso the use of a higher comfort pursuit. the compliance between human and machine is the key factor of comfort when people respiratory.According to the above proposed about the importance of comfort to the ventilator. In this paper, related algorithms are discussed.In this paper, according to the respiratory flow signal acquired by pressure sensor and the clinical results of the statistical analysis for the characteristics of flow signal waveform which provided by ResMed, analyze the characteristics of flow signal waveform. A method of real-time detection of sleep apnea (SA), hypoventilation and obstructive has been proposed. Three kinds of events are Apnea, hypopnea or obstructive. When the system detects any of these events occur during patients sleep, issued a command to uniform increase the motor speed to increased output pressure, until the event disappears. Then stop the boost to achieve the therapeutic effect. When detect the patient in the normal state, the system issued a command to uniform reduce the the motor speed reduce to the output pressure. If an event occurs in the reduction, stoppe reduce. And then increase the motor speed, until the normal state is detected. Repeatedly adjust the size of the output pressure, and ultimately to be an effective treatment adaptive minimum pressure to maximize the use of comfort. System sets the maximum and minimum pressure treatment, the default setting of the maximum therapeutic pressure is10, the minimum pressure of4treatment. You can also independently set under a doctor’s guidance. Monitoring the occurrence of apnea and obstructive is the core of the algorithm. SA detection is to propose a parameter called respiration indices based on the signal characteristics. The size of the respiratory index is consistent with the changes of the fluctuations in the size of the signal. By calculating the index to determine the occurrence of the respiratory event, when the respiration index is less than a threshold value is determined SA event. Obstructive event detection is carried out according to two graphics factor formula to determine the clinical data obtained. Before judgment must first extract the inspiratory wave signal for each breath signal. Then make the inspiratory wave signal as the array into two graphics factor formula to obtain two parameters determine. Simultaneously set two threshold values to judge the two parameters Respectively. When the parameters were within the normal range, it is judged as no obstructive time occurred; if one or two parameters in second in less than a threshold range is determined to obstructive the incident. The method proposed in this paper used by ResMed ventilators and ventilator designed by ourselves to compare experimental results show that this method can effectively monitor the real-time occurrence of apnea, hypopnea and obstructive events in both ventilators, and effectively control the output pressure, and can improve the compliance of the treatment.S AS is a relatively high incidence in the world, therefore the prediction of the condition and evaluation of the therapeutic effect are increasingly attracted people’s attention. Currently, polysomnography is the diagnosis and study of sleep apnea disorders gold standard, also is the most common method for the diagnosis and evaluation of the OSAS. It can not only determine the severity of the disease, but also a comprehensive assessment of the patient’s sleep architecture, sleep apnea, hypoxia, as well as ECG, blood pressure changes. In some cases, central and obstructive can be identified by detecting esophageal pressure. Relying solely on the symptoms described is not sufficient to diagnose. Every patient, should be carried out at least once a PSG examination before treatment, preoperative, postoperative and after treatment. PSG data monitoring checks should be carried out in at least7hours in the sleep lab. PSG testing include EEG, EOG, chin EMG, anterior tibial EMG, ECG, chest and abdominal wall breathing exercises, nose and mouth airflow and oxygen saturation. But the application of PSG for diagnosis and efficacy of OSAS is a complex system engineering, measurement parameters are rich and involve disciplines very extensively, need technical staff and doctor’s diagnosis and treatment. On one hand, it requires detailed, accurate and closely monitored and recorded during sleep in patients. On the other hand, testers must be with a variety of leads so that have a sense of discomfort bundled. And turning the body is likely to cause lead off at night, seriously affecting users sleep, also causing the test results incorrect. The inspection techniques and analysis of PSG are more complex, its cost is relatively high, and the examination is time-consuming. So it is not suitable as a primary means of screening for sleep disorders assessed. Portable monitor PSG selectively uses some parameters, specifically small, portable and easy analysis, has become one of the important ways of screening and diagnosis of OSAS clinic.According to the above proposed about the importance of the portability of portable monitor, in this paper, discuss related algorithms.Pulse is the outside of the information of the heart and major blood vessels state, change in any system of human body will affect pulse. Multiple measurements in surface pulse, such as the wrist, neck, foot wrist, left chest, etc. According to changes in pulse pressure, collected PPG signal waveform will be showing a certain. Radial artery on the wrist acquisition signal is the most universal. The PPG signal of human body is generally considered to have6characteristic points, such as aortic valve opening point, respectivel, the highest systolic pressure point, aortic dilatation buck points, left ventricular ejection stop, reverse tidal wave starting point, dicrotic wave. Based on the characteristics of the PPG signal contains, such as oxygen saturation, pulse rate, pulse transit time, pulse rate, pulse oxygen perfusion index, perfusion index variation, hardening of the arteries and pulse reflection indicators index, variety of human cardiovascular-related physiological and pathological information, calculate related physiological parameters of OSAS patients during sleep with by extracting the characteristic points of PPG signal. Since each of the characteristic point of the waveform is the basis of calculating the above-mentioned physiological parameters. When the wavelet function can be regarded as the second derivative of a smooth function of time, the position relations between the zero-crossing points of wavelet coefficients of the signal and the characteristic points. So take advantage of this feature to identify the characteristic points of the preprocessed signal. Select the smoothed wavelet Marr (Mexican hat wavelet) as the wavelet function. It is because of Marr wavelet basis is the second derivative of Gaussian function, and meet the zero-crossing detection conditions. Meanwhile the wavelet is not compactly supported, not orthogonal, but it is symmetric. So can be used for continuous wavelet transform to obtain a smooth and continuous wavelet coefficients.After decomposition, at the smaller scales, the wavelet transform of the signal amplitude is small. So it is difficult to accurately locate the feature points. And when there is a noise in the signal, the influence of the noise on these scales is serious. While on large scales, position of the feature points would occur offset. In this paper, the offset of position after the wavelet transform is corrected by the differential method. Make differential to extract positions of the first two feature points in the time domain, and subtract the two positions extracted from wavelet transform, and then average to obtain the offset.Particularly, the characteristic points show person dependent features and are easy to be affected. Acquiring a signal with high signal-to-noise ratio (SNR) and integrity is fundamentally important to precisely identify the characteristic points. Based on the mathematical morphologytheory, we design a combined filter, which can effectively suppress the baseline drift and remove the high-frequency noise. We comparatively investigate four typical PWSs reconstructed by three Gaussian functions with tunable parameters. The numerical results suggest that the proposed method can accurately identify the characteristic points of PPG signal.This paper proposes a method of monitoring changes of the patients’ body physiological parameters with sleep apnea syndrome by the PPG signal during sleep, and a method to determine the occurrence of respiratory events and to take treatment using real-time monitoring of respiratory signal. Do an experiment analysis using the sleep apnea signal and PPG signal collected by the acquisition system, the results show that the proposed methods are effective in real time.