Yair Shimoni

Nonlinear filters applied on computerized axial tomography : theory and phantom images

New nonlinear image processing techniques, in particular smoothing based on the understanding of the image, may create computerized tomography (CT) images of good quality using less radiation. Such techniques may be applied before the reconstruction and particularly after it. Current CT scanners use strong linear low-pass filters applied to the CT projections, reducing noise but also deteriorating the resolution of the image. The method in this study was to apply a weak low-pass filter on the projections, to perform the reconstruction, and only then to apply a nonlinear filter on the image. Various kinds of nonlinear filters were investigated based on the fact that the image is approximately piecewise constant. The filters were applied with many values of several parameters and the effects on the spatial resolution and the noise reduction were evaluated. The signal-to-noise ratio of a high-contrast phantom image processed were compared with the nonlinear filter, with the SNR of the phantom images obtained with the built-in CT linear filters in two scanning modes, the normal and the ultra high resolution modes. It was found that the nonlinear filters improve the SNR of the image, compared to the built-in filters, about three times for the normal mode and twice for the UHR scanning mode. The most successful filter on low-contrast phantom image was applied and it also seems to lead to promising results. These results seem to show that applying nonlinear filters on CT images might lead to better image quality than using the current linear filters.(ABSTRACT TRUNCATED AT 250 WORDS)

An Automated System for ST Segment and Arrhythmia Analysis in Exercise Radionuclide Ventriculography

A computer-based system for interpretation of the electrocardiogram (ECG) in the diagnosis of arrhythmia and ST segment abnormality in an exercise system is presented. The system was designed for inclusion in a gamma camera so that ECG diagnosis could be combined with the diagnostic capability of radionuclide ventriculography. Digitized data are analyzed in a beat-by-beat mode and a contextual diagnosis of underlying rhythm is provided. Each beat is assigned a beat code based on a combination of waveform analysis and RR interval measurement. The waveform analysis employs a new correlation coefficient formula which corrects for baseline wander. Selective signal averaging, in which only normal beats are included, is done for an improved signal-to-noise ratio prior to ST segment analysis. Template generation, R wave detection, QRS window size, baseline correction, and continuous updating of heart rate have all been automated. ST level and slope measurements are computed on signal-averaged data. Arrhythmia analysis of 13 passages of abnormal rhythm by computer was found to be correct in 98.4 percent of all beats. 25 passages of exercise data, 1-5 min in length, were evaluated by the cardiologist and found to be in agreement in 95.8 percent in measurements of ST level and 91.7 percent in measurements of ST slope.

Reducing respiratory motion artifacts in nuclear magnetic resonance images

A reordering system for reducing respiratory artifacts in magnetic resonance images. The respiration cycle of the patient is measured and divided into N exclusive intervals. A different encoding amplitude is assigned to each interval in a first look-up table (LUT). The encoding pulses are selected by the first LUT responsive to measured respitory intervals. After M encoding pulses have been selected, a new LUT is constructed from the unused encoding pulse amplitudes using a new number (N-M) intervals. The new LUT is then used for another given number of encoding pulses. Other new LUTs are constructed after given numbers of encoding pulses are used until all encoding pulses are used.

Data predictability for compression of digital fluorography images

Images obtained by digital fluorography were checked for compressability. These images include images of coronary vessels and images of peripheral vessels. These images have a very low signal-to-noise ratio compared to the optical images usually used for developing compression methods. Configurational entropy was used to represent the information content of these images. Reversible prediction algorithms were extensively checked in a search for minimal residual information, enabling more efficient reversible compression. Optimal results were obtained for algorithms based on two or three neighboring pixels and a semiempirical rule, based on the noise level, was found which decides on the best approach. It was found that raw data images are more predictable than subtracted images although the latter are visually preferred.