Angela Oatridge

A registration and interpolation procedure for subvoxel matching of serially acquired MR images

Objective Methods for automatically registering and reslicing MR images using an interpolation function that matches the structure of the image data are described. Materials and Methods Phantom and human brain images were matched by rigid body rotations and translations in two and three dimensions using a least-squares optimization procedure. Subvoxel image shifts were produced with linear or sine interpolation. Results The use of sine interpolation ensured that the repositioned images were faithful to the original data and enabled quantitative intensity comparisons to be made. In humans, image segmentation was vital to avoid extraneous soft tissue changes producing systematic errors in registration. Conclusion The sinc-based interpolation technique enabled serially acquired MR images to be positionally matched to subvoxel accuracy so that small changes in the brain could be distinguished from effects due to misregistration. Index Terms Image registration—Brain—Phantom—Magnetic resonance imaging.

Brain swelling in first hour after coronary artery bypass surgery.

Six patients undergoing routine coronary artery bypass surgery were examined by magnetic resonance imaging of the brain before surgery, immediately afterwards, and 6-18 days later. Brain swelling was visible in all six patients on the immediate postoperative scan. In five patients who had later scans the swelling had subsided. No major neurological deficits were seen, and the patients were extubated successfully within 3 h of the operation. The mechanism of the cerebral swelling is uncertain, but it may provide insight into the cause of neurophysiological deficits seen after coronary artery surgery.

Use of fluid attenuated inversion recovery (FLAIR) pulse sequences in MRI of the brain

Fluid attenuated inversion recovery pulse sequences with a long echo time (TE) have been used to image the brain in one volunteer and four patients. The long inversion time used with this sequence suppresses the signal from CSF and the long TE produces very heavy T2 weighting. The marked reduction in flow artefact from CSF and the high T2 weighting enabled anatomical detail to be seen within the brain stem and produced high lesion contrast in areas close to CSF. Lesions were demonstrated with greater conspicuity than with conventional T2-weighted sequences in patients with cerebral infarction, low grade astrocytoma, and diplegia.

MR imaging of anisotropically restricted diffusion of water in the nervous system: technical, anatomic, and pathologic considerations.

The use of MR imaging to image anisotropically restricted diffusion (ARD) of water in the nervous system is described. The theoretical basis for the use of the pulsed gradient spin echo sequences is outlined, including an estimate of the range of cell dimensions that can be studied with this technique. The importance of restricted diffusion across myelinated white matter fibre tracts is emphasised and the capacity of MR imaging to demonstrate fibre pathways as a function of their direction is illustrated. Technical developments that have been implemented include 256 x 256 spatial resolution, a wider range of diffusion times Td, and an increased range of diffusion sensitivity parameters b. Effects of these are illustrated together with the use of gradient moment nulling methods, oblique sensitisation, and a smaller set of gradient coils that enable shorter values of echo time to be used with the same value of b. The anatomical basis for ARD imaging is analysed, and association, commissural, and projection fibre tracts are demonstrated in different planes. The published literature on variations of the apparent diffusion coefficient from normal is reviewed and examples where diffusion weighted images revealed information that was not necessarily apparent with conventional sequences are illustrated. These include cases of multiple sclerosis, chronic head injury, progressive multifocal leucoencephalopathy, cerebrovascular disease, astrocytoma, and probable metastases to the brain. Imaging of ARD affords a fascinating conjunction between the microscopic movement of water, the properties of myelinated white matter fibres, gross anatomy of the brain, and changes of the diffusion of water in disease.

Detection of subtle brain changes using subvoxel registration and subtraction of serial MR images.

Objective The aim of this study was to examine the potential of accurate image registration for detecting subtle changes in the brain. Materials and Methods Isotropic T1-weighted volume images were obtained in 10 normal subjects and five patients on two or more occasions (including pre- and postcontrast studies). The images were segmented and a 3D rigid body translation and rotation technique was used with sine interpolation to precisely match the images using a x2-test. The registered images and the subtraction images produced from them were used to detect changes in signal intensity, site, shape, and size of the brain. Results Small changes due to differences in orientation of the head, growth, and development as well as inhalation of oxygen and carbogen (95% O2/5% CO2) were observed in normal subjects. Changes were also observed in patients with minor head trauma, a meningioma, an astrocytoma, and multiple sclerosis. Differences due to contrast enhancement and surgery and/or anesthesia were also seen. Conclusion With use of subvoxel registration, subtle changes in the brain were detected in a variety of physiological and clinical situations where differences have hitherto been difficult or impossible to detect.

DESIGN AND IMPLEMENTATION OF MAGNETIZATION TRANSFER PULSE SEQUENCES FOR CLINICAL USE

The transfer of magnetization between a free and a bound pool of spins is described in terms of the respective longitudinal relaxation times and the life times of spins in each pool. The effect of an off resonance radiofrequency (RF) pulse in producing saturation in the bound pool and a consequent decrease in both the available longitudinal magnetization and the T1 of spins in the free pool is described. The effects of increasing duration of the saturating RF pulse on image pixel signal intensity were used to determine values for the decrease in both T1 and the available magnetization in gray and white matter of the brain as well as in muscle, fat, and CSF. At 0.15 T the available magnetization of muscle was reduced by approximately 60% and its T1 was decreased from 350 to 150 ms. The available magnetization of white and gray matter was reduced by 40% and their values of T1 were reduced by 80-110 ms. The reduction in available magnetization was used to increase contrast on proton density weighted or T2-weighted SE pulse sequences. These changes were also used to design inversion recovery (IR) pulse sequences with particular contrast properties. A short inversion time (TI) magnetization transfer (MT) IR (MT-STIR) pulse sequence was used to reduce the signal from normal muscle to zero to produce an angiographic effect in the leg. Increased tissue contrast was observed with a T2-weighted (MT-SE) sequence in a patient with bilateral cerebral infarction and with an MT-IR pulse sequence in a patient who had an intracranial hematoma. Three patients with cerebral tumors showed high lesion contrast with MT-STIR sequences. Components within two tumors were changed to different degrees by MT and in one case change in the brain attributable to recent radiotherapy treatment was only identified with an MT-STIR sequence. Magnetization transfer can be used to manipulate both the available longitudinal magnetization and the T1 of normal and abnormal tissues. The changes in tissue contrast produced by this can be very substantial and are likely to be of importance in clinical imaging.

Cerebral Swelling after Normothermic Cardiopulmonary Bypass

Background Marked cerebral swelling visible on magnetic resonance images has been found immediately after hypothermic (28 [degree sign] Celsius) cardiopulmonary bypass. The mechanism is unknown, but indices of cerebral ischemia are seen during rewarming from hypothermic bypass that are not present with normothermic bypass (37 [degree sign] Celsius). Methods T1‐weighted and fluid‐attenuated inversion recovery magnetic resonance images were taken of seven patients undergoing routine coronary artery bypass surgery before, 1 h, and 7 days after the operation using normothermic bypass. Results Marked cerebral swelling was seen in fluid‐attenuated inversion recovery images in five of seven patients 1 h after bypass. Scans in four patients taken 7 days after bypass showed that the cerebral swelling had returned to normal. There was no change in cerebral ventricular size, and all patients had uncomplicated postoperative courses. Conclusions Normothermic bypass is followed by acute postoperative cerebral swelling. However, the amount of swelling was similar to that found in a previous study after hypothermic bypass. The mechanism of swelling is still obscure, and its relation to neurologic outcome is unknown.

A serial longitudinal quantitative MRI study of cerebral changes in first-episode schizophrenia using image segmentation and subvoxel registration

Lateral ventricular enlargement is the most consistently replicated brain abnormality found in schizophrenia. This article reports a first episode, longitudinal study of ventricular volume using high-resolution serial magnetic resonance imaging (MRI) and recently developed techniques for image registration and quantitation. Baseline and follow-up (on average 8 months later) MRI scans were carried out on 24 patients and 12 controls. Accurate subvoxel registration was performed and subtraction images were produced to reveal areas of regional brain change. Whereas there were no differences between patients and controls with respect to the mean change in ventricular volume, the patients were much more variable in this respect and showed larger increases and decreases. The percentage increase in ventricular size was greater than one standard deviation of control values for 14 patients and the percentage decrease exceeded one standard deviation in eight patients. Although the finding of progressive ventricular enlargement in a proportion of patients supports other studies indicating an ongoing neuropathological process in the early stages of schizophrenia, the reduction of ventricular size in the remaining patients is more difficult to explain. It is suggested that this may reflect improvement in nutrition and hydration following treatment.

MR imaging of the basal ganglia in chronic liver disease: correlation of T1-weighted and magnetisation transfer contrast measurements with liver dysfunction and neuropsychiatric status

Conventional T1-weighted spin echo (T1WSE) and T1-weighted magnetization transfer (MT) images were obtained in 26 patients with biopsy-proven cirrhosis (nine Childs grade A, 10 Childs grade B and seven Childs grade C). Four subjects showed no evidence of neuropsychiatric impairment on clinical, psychometric and electrophysiological testing, seven showed evidence of subclinical hepatic encephalopathy and 15 were classified as having overt hepatic encephalopathy. Signal intensities of basal ganglia nuclei (head of caudate, putamen, globus pallidus and thalamus) and adjacent brain parenchyma were measured and contrast calculated. On T1WSE imaging, contrast measurements of the globus pallidus were significantly greater in patients with neuropsychiatric dysfunction than in those who were unimpaired (p<0.05). This was not observed in the other basal ganglia nuclei. Patients with subclinical and overt hepatic encephalopathy could not be distinguished on the basis of contrast measurements of the globus pallidus or of any other nucleus. T1WSE contrast measurements of the globus pallidus were increased with elevations in blood ammonia levels (p<0.05) and with the severity of liver dysfunction, when graded according to the Pughs score (p<0.05). Those patients with the worst liver injury (Childs grade C) had significantly greater T1WSE pallidal contrast measurements (p<0.05) than those patients with minimal liver injury (Childs grade A). The patients with intermediate liver damage (Childs grade B) could not be distinguished from the other two groups. While MT imaging highlighted the basal ganglia and showed a correlation between globus pallidus contrast and blood ammonia levels (p<0.05), no other relationship between MT contrast measurements and either the degree of hepatic encephalopathy or the severity of liver dysfunction was found.

MR imaging and spectroscopy of the basal ganglia in chronic liver disease: Correlation of T1-weighted contrast measurements with abnormalities in proton and phosphorus-31 MR spectra

The purpose of this study was to correlate the hyperintensity in the globus pallidus seen on T1-weighted magnetic resonance imaging (MRI) of the brain in chronic liver disease with changes in metabolite ratios measured from both proton and phosphorus-31 magnetic resonance spectroscopy (MRS) localised to the basal ganglia. T1-weighted spin echo (T1 WSE) images were obtained in 21 patients with biopsy-proven cirrhosis (nine Childs grade A, eight Childs grade B and four Childs grade C). Four subjects showed no evidence of neuropsychiatric impairment on clinical, psychometric and electrophysiological testing, four showed evidence of subclinical hepatic encephalopathy and 13 had overt hepatic encephalopathy. Signal intensities of the globus pallidus and adjacent brain parenchyma were measured and contrast calculated, which correlated with the severity of the underlying liver disease, when graded according to the Pughs score (p<0.05). Proton MRS of the basal ganglia was performed in 12 patients and 14 healthy volunteers. Peak area ratios of choline (Cho), glutamine and glutamate (Glx) and N-acetylaspartate relative to creatine (Cr) were measured. Significant reductions in mean Cho/Cr and elevations in mean Glx/Cr ratios were observed in the patient population. Phosphorus-31 MRS of the basal ganglia was performed in the remaining nine patients and in 15 healthy volunteers. Peak area ratios of phosphomonoesters (PME), inorganic phosphate, phosphodiesters (PDE) and phosphocreatine relative to BATP (ATP) were then measured. Mean values of PME/ATP and PDE/ATP were significantly lower in the patient population. No correlation was found between the T1WSE MRI contrast measurements of the globus pallidus and the abnormalities in the metabolite ratios measured from either proton or phosphorus-31 MR spectra. Our results suggest that pallidal hyperintensity seen on T1WSE MR imaging of patients with chronic liver disease is not related to the functional abnormalities of the brain observed in hepatic encephalopathy.