Search

BACKGROUND: Obstructive sleep apnoea (OSA) is a common disease that leads to daytime sleepiness and cognitive impairment. Attempts to investigate changes in brain morphology that may underlie these impairments have led to conflicting conclusions. This study was undertaken to aim to resolve this confusion, and determine whether OSA is associated with changes in brain morphology in a large group of patients with OSA, using improved voxel-based morphometry analysis, an automated unbiased method of detecting local changes in brain structure. METHODS: 60 patients with OSA (mean apnoea hypopnoea index 55 (95% CI 48 to 62) events/h, 3 women) and 60 non-apnoeic controls (mean apnoea hypopnoea index 4 (95% CI 3 to 5) events/h, 5 women) were studied. Subjects were imaged using T1-weighted 3-D structural MRI (69 subjects at 1.5 T, 51 subjects at 3 T). Differences in grey matter were investigated in the two groups, controlling for age, sex, site and intracranial volume. Dedicated cerebellar analysis was performed on a subset of 108 scans using a spatially unbiased infratentorial template. RESULTS: Patients with OSA had a reduction in grey matter volume in the right middle temporal gyrus compared with non-apnoeic controls (p

Division of Imaging Sciences, King’s College London, Guy’s Campus, London SE1 9RT, UK, Department of Medical Physics, Guy’s and St Thomas’ NHS Foundation Trust, London SE1 9RT, UK, IMR Unit, St Mary’s Hospital, Praed Street, London W2 1NY, UK, Sir Peter Mansfield Magnetic Resonance Centre, University of Nottingham, University Park, Nottingham NG7 2RD, UK, Centre for Medical Image Computing, New Engineering Building, University College London, Malet Place, London WC1E 6BT, Cancer Research UK Clinical Magnetic Resonance Research Group, Institute of Cancer Research and Royal Marsden NHS Foundation Trust, Downs Road, Sutton SM2 5PT, UK, Department of Radiology, University Hospital, Queen’s Medical Centre, Nottingham NG7 2UH, UK, Radiological Sciences Unit, the Hammersmith Hospitals NHS Trust and Imperial College London, Charing Cross Hospital, London W6 8RF and Optics and Semiconductor Group, Department of Electrical and Electronic Engineering, Imperial College London, Exhibition Row, London SW7 2AT, UK

The absolute signal-to-noise ratio (SNR) measurable from images is defined and related to the theoretical fundamental SNR for MRI. Absolute SNR for quadrature head coils was obtained by both measurement and analysis of published data for MRI systems in the range 0.5-1.5 T. Mean absolute SNR for 22 MRI systems was 1.59 +/- 0.25 x 10(4) Hz1/2 ml-1 T-1 and did not vary significantly with the static field strength or gradient strength and signal bandwidth. This can be related to the fundamental SNR calculated as 2.36 x 10(4) Hz1/2 ml-1 T-1 through an effective noise figure with an average value of 3.4 dB for the 22 systems. Use of this methodology is discussed in relation to acceptance testing and performance assessment.

ed H D Nagel et al COCIR, European Coordination Committee of the Radiological and Electromedical Industries (2000) 85pp, 25 euros This short book was first published in German in 1999. This translation, by one of the contributors (Nagel) and arguably the UK's foremost CT dosimetry expert (Paul Shrimpton), has ensured that the second edition is accurate both in idiom and sense. Although produced in booklet-type binding, the layout is, nevertheless, accessible and the mathematics, diagrams and graphs are clearly set out. The material is organized into six chapters. The introduction contains an historical perspective on CT usage within Europe, but with a particular focus on Germany. This does not detract from its usefulness; indeed it is interesting as many readers will be well aware of the equivalent UK data published from time to time by the NRPB. Chapter two deals with the fundamentals of CT dosimetry, with particular attention upon the various guises of the CTDI (computed tomography dose index). This is a welcome discourse on a subject which often results in confusion. Newer concepts such as the dose length product are introduced and followed through to the definitions of organ and effective dose using the ICRP60 organ factors. Chapter three introduces organ and effective doses, both in terms of practical means of calculation and a review of the literature. Paediatric dose is briefly addressed. Chapter four looks at the theoretical factors which influence dose. It includes a discussion of third versus fourth generation, detector type, spiral and multiple-slice scanners and CT fluoroscopy. It also addresses the related aspect of image quality. The image quality theme is carried into the next chapter which is dedicated to a review of low dose CT in clinical practice. This is a chapter which should have wider appeal to the whole radiological community. Finally there is a chapter on practical measurement procedures with many worked examples. Additionally, there is a very informative appendix on current scanner specifications from seven manufacturers, and a glossary of terms. This is a highly informative book that draws together in one volume the diverse and patchy information on CT dosimetry. It is accurate, up-to-date and accessible. To benefit fully the reader needs to possess some background knowledge of CT technology and methods. My only two minor complaints were that I did not agree with the view that scattered doses in CT fluoroscopy are comparable to conventional fluoroscopy (CT fluoro is substantially higher), and that the referencing system used is rather idiosyncratic. Other than that, this is a super little book. Every medical imaging physicist and engineer should possess a copy. Additionally, parts of it will be useful to the wider radiological community.

Ultrasound-guided aspiration for diagnostic or therapeutic purposes plays an important role in the management of thoracic, abdominal or pelvic fluid collections. The use of a fine needle is safe (Dewbury, 1990) having the advantages of minimal tissue trauma and minimal patient discomfort. The disadvantage of a small needle of 22 gauge is that its bore limits the rate of aspiration, which is significant in dealing with large or viscous collections. The use of a hand held syringe can be tiring for the operator and needle movement or displacement can occur during prolonged aspirations. We describe a simple method of overcoming some of these disadvantages.

The signal values extracted from an NMR image matrix depend upon the NMR properties of the sample material, i.e. the proton density and relaxation times T1 and T2, and also upon various instrument parameters, one such being the slice shape. It is a common approximation to assume an ideal slice profile: one that is perfectly rectangular and invariant in shape for all applications, but in practice real slice profiles are neither. In this study the effect of the slice is included in signal intensity calculations by means of a computer simulation. Comparison is made between these slice-corrected values and values calculated on the basis of an ideal slice. Four pulse sequences are examined: saturation recovery, inversion recovery, spin echo and inversion recovery followed by spin echo.

The quantitative assessment of the imaging performance of nuclear magnetic resonance (NMR) scanning equipment requires the use of test objects (Orr et al, 1983; Lerski et al, 1984). These are commonly constructed from perspex and are filled with aqueous solutions of paramagnetic salts (e.g. CuSO4 or MnCl2) to provide the NMR signal with a relaxation behaviour (T1 in particular) similar to tissue. The presence of the human body during the NMR imaging procedure has the effect of loading the receiver coil. This is due to energy dissipation by currents induced in the conducting tissue. The solutions of paramagnetic salts do not provide conducting pathways and subsequently have very little loading effect. The signal-to-noise ratio is degraded by coil loading and falsely high values will be obtained if measurements are made without matching this effect when using test objects. Saline-filled containers can be used to load receiver coils, but these may be difficult to construct and may interfere with the position...

The quality of images and the accuracy of extracted relaxation time (T1 and T2) values in nuclear magnetic resonance (NMR) imaging is dependent on the characteristics of the slice. It is therefore important that the slice profile can be measured and to know how it behaves under different experimental conditions. Slice shape is determined in conventional selective excitation systems by the spectrum of the radiofrequency pulse used and the nature of the magnetic field gradient which is applied simultaneously. The effectiveness of the selection is also influenced by the subsequent rephasing gradient. Various methods have been used to investigate the slice shape in a practical situation using a 0.1 T resistive magnet NMR scanner and comparisons drawn with the predictions of theory.

A trial was carried out in which the protocols and prototype test objects developed under a European Economic Community concerted research project for the assessment of MR imaging equipment were tested in a series of commercial imaging systems situated throughout Europe. In general, many imperfections were discovered in the imaging performance of the scanners and, in particular, the accuracy and precision with which the relaxation times T1 and T2 could be measured from the images were found to be rather disappointing.

A trial was carried out in which the protocols and prototype test objects developed under a European Economic Community concerted research project for the assessment of MR imaging equipment were tested in a series of commercial imaging systems situated throughout Europe. In general, many imperfections were discovered in the imaging performance of the scanners and, in particular, the accuracy and precision with which the relaxation times T1 and T2 could be measured from the images were found to be rather disappointing.

Application of magnetic resonance imaging (MRI) in two-dimensional quantitative assessment of blood-retinal barrier dysfunction was investigated in rabbits using a 0.1 T (4.25 MHz) resistive system. Reliable and consistent measurements of vitreous T1 were obtained repeatedly, in slices of width 5 mm and X-Y resolution of 1.2 mm. Calibration of reduction of T1 in eyes after injection of gadolinium-DTPA (Gd-DTPA) was performed, resulting in a dose-related response of relaxation rate (1/T1) to the dose of Gd-DTPA injected. Follow-up scans of injected eyes demonstrated a gradual spread of the T1 "hot-spot" as the contrast agent diffused through the vitreous. T1 rose gradually to basal levels by 72 hr. No local effect of Gd-DTPA was found by ophthalmoscopy. Xenon arc photocoagulation of rabbit retina reduced T1 from 1638 +/- 54 (n = 6, mean +/- SD) ms to 1408 +/- 118 (n = 4) msec (P less than 0.01) throughout the vitreous 5-7 hr after treatment. In treated rabbits receiving 1.0 mmol/l Gd-DTPA intravenously, T1 adjacent to lesions 90-120 min after injection was further reduced in a 63 microliter voxel to 670 +/- 50 ms (mean +/- SD, n = 5) with a minimum pixel value of 285 +/- 52 ms. It was estimated that this represented leakage into vitreous of 8.3 nmol Gd-DTPA. Plasma Gd-DTPA concentrations declined rapidly, with half-life of 20-40 min. The findings indicate that MRI is a technique with the potential for repeated quantitative three-dimensional assessment of blood-retinal barrier dysfunction.