Table of contents

Radiobiological research since the time of Lea and Coulson and present day knowledge and work are reviewed.

The ICRU recommends various methods for the energy calibration of electron accelerators in medical use. A simple method, so far rarely used, is based on the Cerenkov effect in gases. The method is described and its practical application discussed. Corrections and error effects on the energy determination are investigated in detail.

A computer-aided method of analysis of the attenuation by known materials of the exposure rate of a narrow beam of X-radiation generated by a constant medium potential at a thick smooth plane target is described. On the assumption that the average loss of energy of initially monoenergetic electrons is given by the Thomson-Whiddington relation, expressions are derived for the differential intensity of radiation leaving the surface of a solid plane target for a number of analytical approximations to the intensity distribution of the bremsstrahlung at its point of origin within the target. Particular forms of these approximations have been determined which yield attenuation curves closely matching those experimentally determined by Thoraeus (1940) for constant exciting potentials of 50, 60, 80 and 100 kV.

As a polychromatic X-ray beam passes through matter, low energy photons are preferentially absorbed, and the (logarithmic) attenuation is no longer a linear function of absorber thickness. This leads to various artifacts in reconstructive tomography. If a water bag is used, the nonlinear attenuation in bone causes a distortion of the bone values and a spill-over inside the skull, or 'pseudo-cortex' artifact. If no water bag is used, there is an additional effect due to the varying thickness of soft tissue which causes a depression of interior values, or 'cupping'. Both artifacts can be remedied by additional prefiltering of the beam and by applying a linearization correction to the detector outputs. These effects have been studied by computer simulation.

Definitions are proposed of line spread and point spread coefficients which may be used to quantify the degree of unsharpness in a radiographic image. Where there are several independent causes of unsharpness the squares of the spread coefficients are additive.

Measurement of cerebral blood flow (CBF) using a ¹³³Xe inhalation technique requires monitoring the clearance rate of the radioisotope from the head using externally situated detectors and also from arterial blood using a detector to monitor the expired air activity. The end-tidal concentration function is assumed to be proportional to the arterial concentration function and this has to be deconvoluted from the clearance functions obtained from the head to enable CBF to be calculated. A procedure using three-dimensional nomograms is derived and tested. It is shown that results obtained using the nomograms do not differ significantly from results obtained using a full deconvolution procedure, and so it is not essential to use a computer.

It is shown that the response for film and the chemical dosemeter does not change significantly up to 42 MeV, while for LiF it decreases by about 10%. Analysis of calculations and measurements showed that this decrease could not be explained by existing theories.

The temperature of the surface of a model of human skin is measured using a recently developed probe to house a thermocouple or thermistor. The results are compared with thermovision and thermodye measurements. The effect on skin temperature of hot and cold vessels at various depths is investigated.

A simple method of determining the absorption corrections for brain function studies is described. All corrections can be calculated in advance and included in the computer program for the image reconstruction applicable to the positron annihilation coincidence detection technique. In order to test the proposed correction scheme, experiments have been carried out with a two-detector system.

When irradiating enzymes with ions, the quasi-elastic damage mechanism becomes dominant at low energies. The incident ions at these energies (<or approximately=2 keV per amu) often lose a significant fraction of their energy even in very thin samples. Therefore, in a comparison with experiment, one needs to average theoretical inactivation cross-sections over the energy degradation process. This involves also considering deflections of the particles from straight-line trajectories in the material. Some simple analytic results which are useful in understanding the measured 'average' inactivation cross-sections are presented.

A simple apparatus including oven and camera is described. The method involves colour-slide photography during heating and the subsequent conversion of the colour of the developed film into the emission spectra using a common absorption spectrometer.

The effect of matched filter smoothing on detectability of abnormalities in radioisotope images was measured by plotting the full receiver operating characteristic (ROC) curve for both smoothed and unsmoothed images.

A system is described which achieves dynamic cardiac imaging by digital data storage without requiring complex instrumentation, employing, with a minimum of modification, data processing facilities already available and being used for analysis and display of gamma camera images. Thus, a conventional ultrasound B-scanner has been interfaced to a Hewlett-Packard HP 2100A computer with a 16k store and both magnetic tape and disc peripherals, and the standard gamma camera software package (HP 5407) has required only minor modifications for application to ultrasonic scanning.

Experimental measurements with a tissue equivalent chamber show that the larger target size of the mixed DT beam neutron generator results in a beam profile in a scattering material quite similar to those obtained with conventional high energy X-ray producing machines if an appropriate collimator is designed to reduce the penumbral width.

The reply by Brahme and Svensson (see ibid., vol.21, p.304 of 1976) to the letter of Atherton and Coleman (see ibid., vol.20, p.658 of 1975) concerning the comparisons of microtron and linear accelerator beams is commented on and some additional data are presented.