Measurement accuracy. Applications in X-ray computed tomography (CT) dose reduction and magnetic resonance spectroscopy (MRS) volume selection

Abstract

X-ray computed tomography (CT) images of transaxial cross sections of the body can be used to determine tissue volumes. However, the considerable radiation dose associated with high-quality clinical images has greatly limited the use of CT for this application. The relationship between the accuracy of the volume measurement and the radiation dose to the patient was analysed. A simple method to evaluate the X-ray-exposure-to-accuracy performance of the CT system was developed and applied. Finally, a manual procedure for radiation exposure control was designed, in which the largest diameter of the patient's cross section is used to determine the CT scan parameters for minimal radiation dose. This procedure was evaluated in clinical routine and resulted in substantially reduced radiation doses with maintained accuracy in the volume determination. Depending on the thickness of the patient, radiation dose levels of 1-45 % of that of a standard CT examination of the abdomen were achieved. This exposure control procedure could be applied to achieve dose reduction in most CT applications. Furthermore, radiation exposure control can and should be a standard feature of every CT system, to allow automatic optimisation of scan parameters for each patient resulting in minimal radiation dose.The potential success of 1H and 31P magnetic resonance spectroscopy (MRS) as a clinical and research tool depends on several factors. One is the ability to accurately restrict the MRS measurement to a tissue volume of interest (e.g. a tumour), despite the fact that the receiver coil of the magnetic resonance (MR) system collects the MR signal from a much larger part of the body. This requires that MRS be used with a volume selection technique, usually combined with MR imaging of the body region to plan the position and size of the MRS volume. A method was developed to experimentally evaluate volume selection performance of the MR system. A new test phantom was designed for point-wise measurements of high precision, and was used to obtain high spatial resolution signal profiles of 31P MRS volume selections with ISIS. Detailed information was obtained of the size, shape and position of the actual MRS volume. Position errors of several mm, signal loss and transition zones of the actual volume in question were demonstrated. The position errors were analysed and significant contributions from off-resonance effects were found, due to measurement parameters that are controllable by the user in both MR imaging and MRS. The spatial inaccuracy, due to magnetic field distortions of the test phantom itself, was analysed and found to be very small, typically a few hundredths of a mm. The mechanical inaccuracy of the phantom was 0.5 mm. The test method developed can be applied to MRS under clinical conditions and requires no special test functionality or adaptation of the MR system. The detailed information provided by this measurement technique can aid the development of more efficient volume selection techniques. It can also be utilised to increase the efficiency of volume-selective MRS in clinical practise.