Search

Introduction: The most commonly used technique of arteriovenous fistula cannulation for hemodialysis is the rope-ladder technique with alternation of the cannulation site. An alternative technique, the buttonhole (BH) technique, has become popular because its constant cannulation sites are advantageous for patients with special characteristics., Objective: To assess the initial experience of our service with the BH technique and determine its usefulness., Patients and Methods: Twenty-one patients with short, tortuous, painful fistulae of difficult cannulation were submitted to the BH technique for the first time using appropriate needles., Results: Neither bleeding nor hematoma were observed during or after hemodialysis. Some patients (15%) reported little or no pain. Two (9.5%) arteriovenous fistulae were lost, and 47.6% of the patients developed fistula clotting at some point in the study, both situations related to change in cannulators. One patient had a paravertebral abscess, possibly originating from the arteriovenous fistula., Discussion and Conclusion: The advantages of reducing pain, miscannulation, and hematoma incidence were counterbalanced by an increased risk of infection and loss of vascular access, due to noncompliance with the technique or involvement of more than one cannulator. The BH technique is useful for selected patients.

A scatter correction algorithm, based on the integral transformation of the projections, has been implemented when using the C.N.R. positron tomograph in Pisa. The performance of the method has been evaluated by measurements on ad hoc phantoms. The technique allows a significant average reduction of the reconstructed scatter fraction. The limitations of this approach have also been investigated.

An interactive software package, performing some useful general purpose image processing operations and being used as a tool for the problem of the anatomical correlation of PET brain images, is under development. The software is developed as a comprehensive tool with a graphic user-interface allowing the display of the processed images through the use of a variety of colormaps. The implemented routines perform a lot of processing operations on the images: a) Local image processing, i.e. smoothing and sharpening, contours extraction, interactive expansion, shrinking and thresholding of the gray scale, histogram equalization. b) ROI handling, i.e., ROI drawing and computing, transformation of an image to a ROI, ROI editing. c) Additional operators include frequency space image processing such as FFT.

This paper describes a procedure proposed for the quantitative evaluation of the influence of the scattering noise on the reconstructed PET images. Applying the same experimental procedure used to evaluate the Scatter Fraction (SF) on the sinograms obtained measuring line sources within a uniform scattering medium, the method here proposed permits the computation, after reconstruction, of a derived index of the SF i.e. the Reconstructed Scatter Fraction (RSF). This offers the advantage of providing an estimate of the scatter noise directly on the images obtained in the same processing conditions as during the in vivo human studies. Comparative data of SF and RSF are reported as well as the effect of varying the filter function and of the measured attenuation correction on the RSF. The preliminary results obtained show that the values of the reconstructed scatter fraction are closer to the SF ones when this last is computed only in the cylinder region and the RSF is computed on images reconstructed with a ramp filter.

Measurement of myocardial blood flow by 13NH3 relies heavily on the assessment of both the input function and the variable tissue extraction fraction. In six open-chest dogs, myocardial and arterial 13NH3 activity was measured both by in vitro sampling and by in vivo positron emission tomography (PET). Regional myocardial blood flow was forced to vary in the range 0.2-5 ml/min/g and actual values were assessed by in vitro counting of 153Gd microspheres. The ammonia input function was processed by: (a) total curve integration; (b) curve integration for 2 min; (c) integral of a fitted curve (gamma variate in vivo and exponential of the downslope in vitro). Method C brought to regional flow values which best approximated microspheres data. The in vitro correlation allows for correcting in vivo values for the flow-dependent extraction fraction. The method can be easily applied for regional myocardial blood flow measurements with PET in human studies.

We describe the characteristics of the CAMIRD/III, a software package for the calculation of radiation dose from internally distributed radionuclides. The work concerns the preparation of a revised version of both CAMIRD/II, which uses prestored data of the specific absorbed-energy fraction, and MIRD-S, which uses the precalculated "S" values. Some improvement in the dose computation has been realized, for instance in the case of organs with walls, The software, written in FORTRAN IV, runs on an IBM 370/168 computer.

This paper describes an investigation of some of the important physical characteristics of a whole-body positron tomograph consisting of two rings of bismuth germanate detectors of dimensions 5.6 mm X 30 mm X 30 mm (512/ring). The resolution applicable to in vivo imaging is six mm or more, depending on radionuclide and reconstruction filter and is very uniform over the field of view normally used. Axial resolution can be varied by moving side collimation (maximum approximately 16 mm FWHM with interplane septa). A simple scheme has been devised to correct for loss of true coincidence events with varying count rate based on the total random and multiple coincidence rates. It is concluded that correction for scattered radiation should be implemented for reliable quantitation.