Biomedical subjects
Publications and source records attributed to G D Barest.
The clinical application of diffusion-weighted and perfusion-weighted magnetic resonance (MR) imaging techniques has revolutionized the diagnosis and management of acute cerebral infarction. This article discusses the physical basis of these advanced techniques and briefly demonstrates their application along with standard MR imaging and MR angiography in the setting of acute stroke.
The objective of radiation therapy treatment is to eradicate a cancerous tumor while keeping the damage to nearby healthy organs to a minimum. A variety of tools employing computer graphics exist to aid in the planning and verification of treatments. Three-dimensional (3D) image information available from sources such as computerized tomography (CT) scanners is used to define the sizes, shapes, and spatial locations of the tumor and normal structures in the form of transverse contours. These object definitions are displayed in 3D perspective to enable the determination of the best possible directions from which to aim radiation beams at the tumor. The beams may be shaped to match the outline of the tumor, and their intensities may be modified using compensating devices. The results of calculations done to predict the distribution of radiation dose throughout the body due to a given set-up of beams can be displayed to the user in many ways. Dose may be shown in the form of isodose contours overlaid on transverse CT images, or on reconstructed image planes of arbitrary orientation in space. There are also a number of methods of 3D display; dose can be shown on the surface of objects, or in the form of isodose surfaces relative to anatomical structures. Computer-generated beam film images may be used to verify patient set-up and tumor coverage.
Graphical displays of three-dimensional dose distribution data are often too complex to be easily assimilated and interpreted for the evaluation of radiation treatment plans. Histograms showing dose versus volume are convenient and useful tools for summarizing dose distribution information throughout the entire volume of a given anatomic structure. They can quickly highlight characteristics such as dose uniformity and hot and cold spots, and can be used to produce statistics including tumor control and normal tissue complication probabilities. To obtain a dose volume histogram for a given structure, it may be necessary to examine its spatial relationships with neighboring structures. They may overlap, be completely disjoint, or one may be contained within another. To resolve potential ambiguities, a procedure has been developed that assigns hierarchies to anatomical structures for the purpose of histogram calculation. The hierarchy assigned to each structure is used to determine the structure within which a given dose matrix point is considered to lie. In this manner, regions of structure intersection are assigned to one object or another, and dose volume histograms can be calculated for each structure separately. From this framework, addition and subtraction of histograms can also be performed. Details of the algorithm are presented along with an example using patient data.
Presented here is a method of designing compensators for a single beam or one or more pairs of beams, not necessarily parallel opposed. The objective is to produce a flat distribution in a plane that may be perpendicular to the central ray or may be an arbitrarily oriented plane, for example, a plane that bisects the hinge angle between two beams. The method takes into account not only surface irregularities but also tissue inhomogeneities, hinge angles between beams, distance from the source, and even "horns" in the beam. The design process employs convolution of Monte Carlo generated pencil beams with photon fluence distributions, appropriately modified for the presence of beam modifiers (blocks and compensators), to compute dose in a flat homogeneous phantom. Corrections for inhomogeneities and surface curvature are applied by using computerized tomography information to determine the effective path length through tissue. Multiple interactions are used to arrive at a compensator that properly incorporates changes in radiation transport, and therefore dose distribution, resulting from the presence of beam-shaping devices. In each iteration it is assumed that the required reduction in dose at a point can be achieved by reducing the fluence along the ray joining the source to computation point proportionately. The compensator design is represented as a finely spaced matrix of thickness values which is entered into a prorammable milling maching for fabrication. Dose measurements in phantom exposed to 6-MV x rays with and without compensation are presented.
INTRODUCTION: In the evaluation of stenoses of the extracranial internal carotid artery (ICA), there are studies that suggest that magnetic resonance angiography (MRA) can be a substitute for conventional arteriography (CA), although it seems it has a tendency to overestimate the degree of stenosis. No similar comparison of the two techniques has been conducted in intracranial ICA. We report the case of a patient suffering from an acute ischemic stroke and symptomatic intracranial stenosis that was overestimated when MRA was used, compared to the results obtained using CA. CASE REPORT: We report the case of a 64-year-old male with a history of arterial hypertension, hypercholesterolemia and intermittent claudication who visited the emergency department because of the sudden onset of paresthesias in the left hemiface and hand. The cranial tomography scan performed in the emergency unit ruled out any acute bleeding or early signs of a stroke. Magnetic resonance (MR) diffusion imaging showed an acute ischemic stroke in the right parietal cortex. Extracranial MRA was normal and in the intracranial area a 73% stenosis was detected in the cavernous segment of the right ICA, whereas the use of CA showed the stenosis to be only 55%. On repeating the MRA to rule out a possible rechanneling of the ICA, the image obtained was exactly the same as the earlier one. CONCLUSIONS: Our observations suggest that, as occurs with the extracranial part, MRA tends to magnify the degree of stenosis in the intracranial vessels, and this technique would therefore appear to be less efficient than CA in the evaluation of intracranial stenoses.