Computers and biomedical research, an international journal最新文献

A real-time analysis (RTA) system, based on a personal computer with a digital signal processor card (DSP) was developed. The system extracts and saves the QT and RR intervals from an incoming ECG signal sampled at 1 kHz. The method of defining the QT and RR intervals is based on performing multiple cross-correlations that enables rejection of artifacts from the analysis. The relationship between the RR and the QT intervals is found using the following general formula QT_icRRⁱ_i−1. Linear regression is performed on the logarithms of QT and RR measurements obtained to estimate a constant (a = logc) and a slope (b), reflecting the dynamic change of the QT intervals. Having these two values, the dynamic QT extrapolated to a heart period of 1 s (QTcd) was calculated. The system also performs spectral analysis of the intervals and dynamic QT-RR relation analysis. The system was evaluated on recordings from 10 infants aged 4 to 24 weeks. The QT/RR computerized measuring device used in the present study answers to the requirements for precise dynamic beat-to-beat QT measurement. Its sampling rate is high and can achieve a real millisecond-precision without the need of interpolation methods. The measurements are performed on-line, at the time of the actual recording. Results appear immediately and a measure of dynamic QT behavior can be easily obtained within minutes. The computerized system with the dynamic QT-RR measurements may provide a simple and accurate tool for testing drug treatment and effects of other interventions for cardiac disorders and arrhythmia risk.

A lumped two-compartment mathematical model of respiratory mechanics incorporating gas exchange and pulmonary circulation is utilized to analyze the effects of square, descending and ascending inspiratory flow waveforms during mechanical ventilation. The effects on alveolar volume variation, alveolar pressure, airway pressure, gas exchange rate, and expired gas species concentration are evaluated. Advantages in ventilation employing a certain inspiratory flow profile are offset by corresponding reduction in perfusion rates, leading to marginal effects on net gas exchange rates. The descending profile provides better CO₂ exchange, whereas the ascending profile is more advantageous for O₂ exchange. Regional disparities in airway/lung properties create maldistribution of ventilation and a concomitant inequality in regional alveolar gas composition and gas exchange rates. When minute ventilation is maintained constant, for identical time constant disparities, inequalities in compliance yield pronounced effects on net gas exchange rates at low frequencies, whereas the adverse effects of inequalities in resistance are more pronounced at higher frequencies. Reduction in expiratory air flow (via increased airway resistance) reduces the magnitude of upstroke slope of capnogram and oxigram time courses without significantly affecting end-tidal expired gas compositions, whereas alterations in mechanical factors that result in increased gas exchanges rates yield increases in CO₂ and decreases in O₂ end-tidal composition values. The model provides a template for assessing the dynamics of cardiopulmonary interactions during mechanical ventilation by combining concurrent descriptions of ventilation, capillary perfusion, and gas exchange.

In this work we have implemented a system for the automatic segmentation of lung fields in chest radiographic images. The image analysis process is carried out in three levels. In the first one we perform operations on the image that are independent from domain knowledge. This knowledge is implicitly and not very elaborately used in the intermediate level and used in an explicit manner in the high level block, globally corresponding to the idea of progressive segmentation. The representation of knowledge in the high level block is in the form of production rules. The control structure is in general bottom-up but there are certain hybrid control stages, in which the control is driven by the region model (main organs) we are seeking. We have applied the global system to a set of 45 posteroanterior (PA) chest radiographs, obtaining a mean degree of overlap with contours drawn by radiologists of 87%.

Comparison of incidence and mortality data between areas is useful for monitoring the health status of a population, for exploring quality of health care, and for planning studies on determinant(s) of heterogeneous patterns. Common statistical indicators involved, together with their confidence intervals, are age-specific and cumulative rates, age-standardized rates and ratios, and expected numbers. The data required for calculations include information of event distribution according to age and residence, and corresponding population at risk. The calcula tions of these indicators, although conceptually simple, can be a cumbersome task to carry out with available statistical packages; lack of flexibility to suit the need of specific researches represents another problem. In order to facilitate descriptive epidemiological studies, we devel oped a set of Statistical Analysis System macros to compute indicators but, also, to read and write data from/to some standard file formats. Important features of these macros are their flexibility, expandibility, and ability to be transferred to various computer platforms. Moreover, macros give an added value to raw data. The method is illustrated using published incidence data from Italy.

Conventional brain maps suffer from severe limitations due to both the spatial blur of potential distributions and the dependence on electrical reference. The surface Laplacian (SL) has been used to deblur movement-related brain macropotentials (MRBM) since it acts as a high-pass spatial filter that reduces the head volume conductor effects. Moreover, the method usually employed to improve the signal-to-noise ratio (SNR) is the well-known synchronized average. However, this method is no longer valid when the object of the study is the sweep-by-sweep variability. In this case, the SNR of original and Laplacian-transformed single-sweep MRBM can be improved by autoregressive with exogenous input (ARX) filtering. In our study, isolated or combined ARX and SL are applied to enhance the spatial distributions of single-sweep MRBM associated with unilateral voluntary self-paced finger movements in humans. It shows that single-sweep brain mappings are more coherent to physiological findings when ARX is first used followed by SL.

The geometry of the heart plays a major role in cardiac function. The purpose of this study was to characterize analytically the geometric properties of the left ventricular (LV) three-dimensional (3D) shape, while excluding the effects of aspect ratio and size. Two groups of human hearts were studied by Cine-CT. The first group was composed of 10 healthy volunteers and the second of 9 pathological hearts. The hearts were scanned from apex to base. The endocardial borders of each LV scan were traced and used to reconstruct the 3D LV at end-diastole (ED) and end-systole (ES). Using a special normalized helical shape descriptor, denoted “geometrical cardiogram” (GCG), the typical 3D normal ED and ES shapes were determined. These typical shapes were then analytically approximated via a discrete cosine transform (DCT). The shape of each LV was then investigated for its correspondence to five analytically defined shapes: (i) a cone, (ii) a sphere, including all ellipsoidal shapes, (iii) a cylinder, (iv) a truncated ellipsoid, and (v) the DCT approximation of the normal LV shape. The results indicate that the normal LV shape can be well approximated by using only seven coefficients of the DCT. Conicity was the only geometrical feature which did not change from ED to ES in the normal group of hearts. The most prominent shape difference between normal and abnormal hearts was the significantly reduced conicity of the latter. Conicity is an important feature of LV geometry. The possible contribution of the conical shape to LV ejection efficiency is also discussed.

Changes in onset latency and relative amplitudes of somatosensory evoked potentials (SEP) may be a convenient and reliable neurophysiological indicator of depth of anesthesia. However, to derive the components is very difficult mathematically and visual inspection or alternatively the peak-latency estimation is usually employed. A methodology for estimating the components was developed for both real-time and off-line applications based on the combination of the wavelet transforms (WT), geometric analysis, artificial intelligence (AI), and mathematical analysis of the first positive wave of SEPs. The WT together with AI constitutes a feature extraction engine for localizing the first positive peak and negative valley and hence relative amplitudes. The latency change between two averages is obtained by shifting one average toward another to achieve a best match along the positive inflections. The inflection, based on the peak, is modeled as a regression line and is refined using a steepness inference algorithm. Results from simulation and anesthetized rats show that it is reliable in comparison with visual inspection, robust to amplitude variation and signal distortion, and efficient in computation, and hence it is suitable for automation. Comparisons of interobserver variability and analysis of method agreement suggest that the method can be used as a substitute for estimations by visual inspection.

Self-learning fuzzy logic control has the important property of accommodating uncertain, nonlinear, and time-varying process characteristics. This intelligent control scheme starts with no fuzzy control rules and learns how to control each process presented to it in real time without the need for detailed process modeling. In this study we utilize temporal knowledge of generated rules to improve control performance. A suitable medical application to investigate this control strategy is atracurium-induced neuromuscular block of patients in the operating theater where the patient response exhibits high nonlinearity and individual patient dose require ments may vary fivefold during an operating procedure. We developed a computer control system utilizing Relaxograph (Datex) measurements to assess the clinical performance of a self-learning fuzzy controller in this application. Using a T1 setpoint of 10% of baseline in 10 patients undergoing general surgery, we found a mean T1 error of 0.28% (SD = 0.39%) while accommodating a 0.25 to 0.38 mg/kg/h range in the mean atracurium infusion rate. This result compares favorably with more complex and computationally intensive model-based control strategies for atracurium infusion.

According to Fowler's method, anatomical dead space (V_D) can be determined graphically or computer-aided by iteration procedures by which phase III of a fraction–volume expirogramF(V) is back-extrapolated by a straight lineR(V). Whereas Fowler visually partitioned phase II into two equal areas bordered byF(V),R(V), andV_D, in the present paper the area betweenF(V) andR(V) is set equal to the area of a trapezoid, one side of which is the unknownV_Dto be determined. We obtained two algebraic equations for both possible conditions, nonsloping and sloping alveolar plateau, and, as the main result, an even more general third equation that includes both Bohr's and Fowler's solution. The formulas exactly represent Fowler's graphical method and can be applied to all gases which are applicable in dead space determination. The derived equations were tested in experimental situations, showing equality between values of dead space determined by using the algebraic solution and the graphical method. Their major advantage is facilitating and speeding up computer-aided on-line determinations ofV_D.