Comparisons of extra-cellular fluid (ECF) volume estimates made by isotope dilution and electrical impedance techniques have been made in a group of 16 children. For each patient an estimate of ECF volume (Vt) was obtained from measurements of the blood clearance of 99Tcm-diethylene triamine penta-acetate (DPTA) which was compared with two estimates (Vi1 and Vi2) of ECF volume obtained from measurements of the whole-body electrical impedance at 50 frequencies in the range 1 kHz to 1.348 MHz and a third estimate Vh based on patient height, L, alone. The observed frequency response of the impedance measurements was fitted to a three-element equivalent-circuit model of whole-body impedance and gave a value of the ECF resistance R. Vi1 was obtained from Vi1 = a (L2/R) + b. Vi2 was given by c(W1/2L2/R)2/3 where W is the patient weight, and Vh was given by dL2 + e. The constants a, b, c, d, e were determined by comparison with Vt and were equal to 0.335 l omega m-2 (standard error = 0.01 1 omega m-2), 0.42 l (0.25 l), 0.33 l (omega 2kg-1m-4)1/2 0.007 l (omega 2kg-1m-4)1/3, 4.92 l m-2 (2.8 x 10(-5) lm-2), 0.13 l (0.41 l), respectively. Vi1, Vi2, Vh were linearly correlated with Vt (r2 = 0.98, 0.99, 0.95, respectively, p < 0.001), and upper and lower levels of agreement were given by +/- 0.95 l (Vt and Vi1), 1.44 l and -1.12 l (Vt and Vi2), +/- 1.5 l (Vt and Vh), respectively. Thus inclusion of the impedance data accounted for greater volume variation, but differences between the techniques were not significant (paired t-test and Mann-Whitney analysis) suggesting that more accurate and detailed measurements are required.

The paper describes an approach to solving the problem of providing a large-capacity image archive for diagnostic imaging departments at reasonable cost. Optical disk stores, when fitted retrospectively to scanners, are very expensive and may not be compatible with existing computer hardware. We describe the use of an industry standard personal computer (PC) linked to a standard 5 1/4-in. optical disk drive as a 'stand-alone' image store. Image data are transferred from the scanner using 8-in. floppy disks, and these are read into the PC using an attached 8-in. floppy disk drive and then transferred to the optical disk. The patient details (patient name, ID, date, etc.) are entered into a database program held on the PC and these are used to generate a reference pointer to the optical disk file through which the data can be retrieved. Data retrieval involves entering the patient details into the data base and inserting a blank 8-in. floppy disk into the drive attached to the PC. A sector copy is then made from the optical disk to the 8-in. floppy disk, which can then be used at the viewing station at the scanner. The system is flexible since it can accept data from a variety of sources in any format; it is also low cost and operates independently of the scanner. The hardware is industry standard, ensuring low maintenance costs.

The operation of a commercially available nebuliser system (Medic-Aid Ltd) is reviewed and the efficiency with which it produces an aerosol assessed. Defining the efficiency of nebulisation E as the fraction of the original mass of solution released as an aerosol it is found that the internal surface area, mass of solution used, the surface tension of the solution and the angle of tilt are important factors in determining E. Reducing the internal surface area of the nebuliser by means of Perspex inserts significantly increases E for 3 g of water from 49% for the unmodified system to 67% for the modified nebuliser (P less than 0.01). E increases with the mass of solution used but only exceeds 60% when 4.5 g water are used. Decreasing the surface tension of the solution from 7.2 x 10(-3) N m-1 (water) to 3.7 x 10(-3) N m-1 and 3.1 x 10(-3) N m-1 (using two different concentrations of a detergent in water) significantly increases E for 3 g solution from 49% to 65% and 69% respectively (P less than 0.05). Operating the nebuliser at a tilt also increases E. The measurements emphasise the importance of reducing the internal surface area of this type of nebuliser, using an adequate volume of drug solution (at least 4 ml is suggested) and operating the nebuliser at an angle to the vertical (20 degrees suggested) in order to maximise E. The surface tension of the drug solution is a further important determinant of E.

A system using closed-circuit television has been developed to enable the measurement of patient movements to an accuracy of 1 mm during external beam radiotherapy. The circuitry is under the control of a microcomputer, which records the movement data on magnetic tape for later analysis. The results indicate that patients mostly remain within 1 mm of their original set-up position on the treatment couch.

The intrinsic spatial resolution of a gamma camera is specified by the full width at half maximum (FWHM) of the line spread function (LSF), in the absence of a collimator, and is an important indicator of system performance. The standard method of determining the spatial resolution is to place a suitably collimated line source against the crystal face and to examine the amplitude spectrum from the camera. For a modern gamma camera with a field of view of 375 mm, the expected FWHM may be 4 mm; to measure this accurately would require a sampling frequency of 0.4 mm. Most nuclear medicine computers are able to acquire data with a maximum resolution of 256×256 pixels per frame, corresponding to a sampling frequency of about 1.5 mm, which is too coarse for measurement of the LSF. A low-cost 1024-channel multichannel analyser (MCA) for γ-ray spectrometry, interfaced to a personal computer, can be adapted to measure the LSF of the camera and calculate the FWHM with sufficient accuracy (sampling frequency better than 0.4 mm). The equipment to be described has been used on three gamma cameras in routine use.