基于小波变换和似然无偏估计的胃阻抗信号处理研究

目的：介绍一种基于小波变换和似然无偏估计的胃阻抗信号分析去噪的方法。方法：首先经过多分辨分析将信号进行分离，然后应用Stein似然无偏估计对胃阻抗信号进行自适应软阈值去噪处理。结果：实验结果表明，该方法能够有效滤除由呼吸、血流阻抗等因素造成的干扰。结论：小波分析技术为进一步研究胃动力信号的参数分析提供了新途径。     

The electrocardiogram in acute pulmonary embolism

Electrocardiograms of 90 patients with arteriographically documented acute submassive or massive pulmonary embolism and no associated cardiac or pulmonary disease were studied. Patients were derived from the Urokinase-Pulmonary Embolism Trial National Cooperative Study. In massive embolism, the electrocardiogram was normal in 6% (3 of 50) of patients. With submassive embolism, 23% of patients (9 of 40) had a normal electrocardiogram. Since one or more of the traditional manifestations of acute cor pulmonale (S₁Q₃T₃, right bundle branch block, P pulmonale, or right axis deviation) occurred in only 26% of patients, one could not rely exclusively upon these electrocardiographic abnormalities for the diagnosis of pulmonary embolism. The most common electrocardiographic abnormalities were nonspecific T wave changes which occurred in 42% of patients and nonspecific abnormalities (elevation or depression) of the RST segment which occurred in 41% of patients. Left axis deviation occurring in 7% of the patients was as frequent as right axis deviation. Low voltage QRS complexes, previously undescribed in pulmonary embolism, occurred in 6% of patients. None of the patients had atrial flutter or atrial fibrillation, which appears to occur more typically in patients with pulmonary embolism who have preexistent cardiac disease. All of the varieties of electrocardiographic abnormalities disappeared in some of the patients by 2 wk. Inversion of the T wave was the most persistent abnormality. Larger defects on the lung scan or pulmonary arteriogram occurred in patients with various abnormalities on the electrocardiogram than in patients with normal electrocardiograms. The pulmonary arterial mean pressure and/or right ventricular end-diastolic pressure was significantly higher in patients with several varieties of abnormal electrocardiograms, although the partial pressure of oxygen in arterial blood, in general, did not differ from that in patients with normal electrocardiograms. These hemodynamic correlations, made for the first time in patients, suggest that acute ventricular dilatation, possibly in combination with hypoxemia, is a causative factor of the electrocardiographic changes in acute massive or submassive pulmonary embolism.     

Determinants of the amplitude of the aortic component of the second heart sound in aortic stenosis

The characteristics of the aortic component of the second heart sound in calcific and congenital noncalcific aortic stenosis were studied to determine a cause for observed differences. Intraarterial pressure and sound were measured above the aortic valve in 20 patients utilizing catheter-tip micromanometers. Ten patients had a normally functioning aortic valve, six had calcific aortic stenosis and four had congenital noncalcific aortic stenosis. As expected, the aortic sound was diminished in patients with calcific aortic stenosis compared with that in patients with a normal valve (600 ± 200 versus 2,600 ± 200 dynes/cm² (P < 0.001). In patients with congenital aortic stenosis, sound amplitude was not reduced compared with that in patients with a normal valve. Measurement of sound produced by closure of normal and stenotic valves in an in vitro model of the circulatory system yielded comparable results. In vitro high speed (2,000 frames/sec) motion pictures of the diastolic motion of the closed cusps showed vibrations of comparable magnitude in the normal porcine and the simulated congenitally stenotic valve. The calcified stenotic valve showed no noticeable diastolic vibrations. These observations indicate an association between the amplitude of the second heart sound and diastolic vibrations of the closed cusps. A calcified stenotic valve, being thick and stiff, would have a diminished ability to vibrate and would therefore produce a diminished sound. A congenitally stenotic valve, in contradistinction, if not yet damaged by degenerative changes, would not be limited in its ability to vibrate during diastole and would therefore produce a normal second sound.     

Performance of the failing and nonfailing right ventricle of patients with pulmonary hypertension.

Hemodynamic performance of the right ventricle was measured in 34 patients: 17 with pulmonary hypertension, 9 with pulmonary hypertension and right ventricular failure and 8 control subjects. Among the patients with pulmonary hypertension who did not have right ventricular failure, right ventricular maximal isovolumic rate of development of ventricular pressure (dP/dt) was significantly elevated (P less than 0.001), whereas maximal 1/P dP/dt and maximal velocity of contractile element shortening (Vmax) were comparable with values observed in control subjects. The patients with pulmonary hypertension who had right ventricular failure also showed an augmented right ventricular maximal dP/dt (P less than 0.001) and normal 1/P dP/dt and Vmax. These observations indicate that in pulmonary hypertensive heart disease, even when the right ventricle failed in a clinical sense, the contractile effort was normal. Consequently, right ventricular failure may develop in patients with pulmonary hypertensive heart disease even though the cardiac muscle performs normally as a contractile tissue.     

Ventricular performance in patients based upon rate of change of power during isovolumic contraction

Energy is expended by the ventricle during isovolumic contraction as the blood within the ventricle is compressed. This isovolumic energy of compression, as well as the isovolumic rate of energy transfer (power) and acceleration of energy transfer (rate of change of power), was calculated in 17 patients with angina pectoris who underwent diagnostic cardiac catheterization. The peak isovolumic rate of change of power in patients with normal left ventricular performance (based upon the ejection fraction, mean velocity of circumferential fiber shortening and end-diastolic volume index) was 31,000 ± 3,000 dynes cm sec^?2 (mean ± standard error), whereas in those with poor ventricular performance it was 18,000 ± 2,000 dynes cm sec^?2 (P < 0.01). None of the patients with poor ventricular performance had a peak isovolumic rate of change of power that exceeded 25,000 dynes cm sec^?2. Overlap between patients with normal performance and those with reduced performance was less with the peak rate of change of power than with peak power, peak rate of change of intraventricular pressure ($\text{dp}\text{dt}$), maximal (dp/dt)/p or Vmax. The derivation of these isovolumic energy transfer rates requires no assumptions related to ventricular geometry or characteristics of muscle fibers. Peak isovolumic rate of change of power may be a useful and sensitive indicator of ventricular performance in patients with coronary artery disease.