High-throughput, accurate Monte Carlo simulation on CPU hardware for PET applications.

Monte Carlo simulations (MCS) represent a fundamental approach to modelling the photon interactions in positron emission tomography (PET). A variety of PET-dedicated MCS tools are available to assist and improve PET imaging applications. Of these, GATE has evolved into one of the most popular software for PET MCS because of its accuracy and flexibility. However, simulations are extremely time-consuming. The use of graphics processing units (GPU) has been proposed as a solution to this, with reported acceleration factors about 400-800. These factors refer to GATE benchmarks performed on a single CPU core. Consequently, CPU-based MCS can also be easily accelerated by one order of magnitude or beyond when exploiting multi-threading on powerful CPUs. Thus, CPU-based implementations become competitive when further optimisations can be achieved. In this context, we have developed a novel, CPU-based software called the PET physics simulator (PPS), which combines several efficient methods to significantly boost the performance. PPS flexibly applies GEANT4 cross-sections as a pre-calculated database, thus obtaining results equivalent to GATE. This is demonstrated for an elaborated PET scanner with 3-layer block detectors. All code optimisations yield an acceleration factor of ≈20 (single core). Multi-threading on a high-end CPU workstation (96 cores) further accelerates the PPS by a factor of 80. This results in a total speed-up factor of ≈1600, which outperforms comparable GPU-based MCS by a factor of ≳2. Optionally, the proposed method of coincidence multiplexing can further enhance the throughput by an additional factor of ≈15. The combination of all optimisations corresponds to an acceleration factor of ≈24 000. In this way, the PPS can simulate complex PET detector systems with an effective throughput of 106photon pairs in less than 10 milliseconds.

Technical Note: GATE-RTion: a GATE/Geant4 release for clinical applications in scanned ion beam therapy.

PURPOSE: GATE-RTion is a validated version of GATE for clinical use in the field of light ion beam therapy. This paper describes the GATE-RTion project and illustrates its potential through clinical applications developed in three European centers delivering scanned proton and carbon ion treatments. METHODS: GATE-RTion is a collaborative framework provided by the OpenGATE collaboration. It contains a validated GATE release based on a specific Geant4 version, a set of tools to integrate GATE into a clinical environment and a network for clinical users. RESULTS: Three applications are presented: Proton radiography at the Centre Antoine Lacassagne (Nice, France); Independent dose calculation for proton therapy at the Christie NHS Foundation Trust (Manchester, UK); Independent dose calculation for protons and carbon ions at the MedAustron Ion Therapy center (Wiener Neustadt, Austria). CONCLUSIONS: GATE-RTion builds the bridge between researchers and clinical users from the OpenGATE collaboration in the field of Light Ion Beam Therapy. The applications presented in three European facilities using three completely different machines (three different vendors, cyclotron- and synchrotron-based systems, protons, and carbon ions) demonstrate the relevance and versatility of this project.

Multi-modality imaging of uveal melanomas using combined PET/CT, high-resolution PET and MR imaging.

UNLABELLED: We investigated the efficacy of combined FDG-PET/CT imaging for the diagnosis of small-size uveal melanomas and the feasibility of combining separate, high-resolution (HR) FDG-PET with MRI for its improved localization and detection. PATIENTS, METHODS: 3 patients with small-size uveal melanomas (0.2-1.5 ml) were imaged on a combined whole-body PET/CT, a HR brain-PET, and a 1.5 T MRI. Static, contrast-enhanced FDG-PET/CT imaging was performed of head and torso with CT contrast enhancement. HR PET imaging was performed in dynamic mode 0-180 min post-injection of FDG. MRI imaging was performed using a high-resolution small-loop-coil placed over the eye in question with T2-3D-TSE and T1-3D-SE with 18 ml Gd-contrast. Patients had their eyes shaded during the scans. Lesion visibility on high-resolution FDG-PET images was graded for confidence: 1: none, 2: suggestive, 3: clear. Mean tumour activity was calculated for summed image frames that resulted in confidence grades 2 and 3. Whole-body FDG-PET/CT images were reviewed for lesions. PET-MRI and PET/CT-MRI images of the head were co-registered for potentially improved lesion delineation. RESULTS: Whole-body FDG-PET/CT images of 3/3 patients were positive for uveal melanomas and negative for disseminated disease. HR FDG-PET was positive already in the early time frames. One patient exhibited rising tumour activity with increasing uptake time on FDG-PET. MRI images of the eye were co-registered successfully to FDG-PET/CT using a manual alignment approach. CONCLUSIONS: Small-size uveal melanomas can be detected with whole-body FDG-PET/CT. This feasibility study suggests the exploration of HR FDG-PET in order to provide additional diagnostic information on patients with uveal melanomas. First results support extended uptake times and high-sensitivity PET for improved tumour visibility. MRI/PET co-registration is feasible and provides correlated functional and anatomical information that may support alternative therapy regimens.

Does PET/CT render software registration obsolete?

It was the success of software-based image registration that eventually led to the introduction of hardware-based concepts for image fusion, such as combined PET/CT tomographs. A prototype PET/CT was first presented in 1998, with various commercial designs to follow since 2000. PET/ CT is used primarily as a diagnostic modality in the field of extra-cerebral oncology imaging. The major advantage of combined imaging over retrospective software registration is the nearly identical position of the patient during both complementary examination, and therefore tomograms of identical parts of the body can be provided in spatially-corresponding slices. Despite the availability of hardware combinations of complementary imaging modalities software-based image registration, however, still inherits a major role in subsequent data processing, in particular when individual imaging modalities other than combined PET/CT are being used during patient workup. Furthermore, software is likely to become an important tool for the correction of residual motion-induced mis-registration within combined PET/CT data sets, and for follow-up studies involving, for example, CT, PET, and PET/CT. Therefore, flexible algorithms that utilize non-linear interpolation schemes implemented on fast computer systems are needed, and will continue to contribute to successful image registration and fusion in clinical practice.

High performance volume-of-intersection projectors for 3D-PET image reconstruction based on polar symmetries and SIMD vectorisation.

For high-resolution, iterative 3D PET image reconstruction the efficient implementation of forward-backward projectors is essential to minimise the calculation time. Mathematically, the projectors are summarised as a system response matrix (SRM) whose elements define the contribution of image voxels to lines-of-response (LORs). In fact, the SRM easily comprises billions of non-zero matrix elements to evaluate the tremendous number of LORs as provided by state-of-the-art PET scanners. Hence, the performance of iterative algorithms, e.g. maximum-likelihood-expectation-maximisation (MLEM), suffers from severe computational problems due to the intensive memory access and huge number of floating point operations. Here, symmetries occupy a key role in terms of efficient implementation. They reduce the amount of independent SRM elements, thus allowing for a significant matrix compression according to the number of exploitable symmetries. With our previous work, the PET REconstruction Software TOolkit (PRESTO), very high compression factors (>300) are demonstrated by using specific non-Cartesian voxel patterns involving discrete polar symmetries. In this way, a pre-calculated memory-resident SRM using complex volume-of-intersection calculations can be achieved. However, our original ray-driven implementation suffers from addressing voxels, projection data and SRM elements in disfavoured memory access patterns. As a consequence, a rather limited numerical throughput is observed due to the massive waste of memory bandwidth and inefficient usage of cache respectively. In this work, an advantageous symmetry-driven evaluation of the forward-backward projectors is proposed to overcome these inefficiencies. The polar symmetries applied in PRESTO suggest a novel organisation of image data and LOR projection data in memory to enable an efficient single instruction multiple data vectorisation, i.e. simultaneous use of any SRM element for symmetric LORs. In addition, the calculation time is further reduced by using simultaneous multi-threading (SMT). A global speedup factor of 11 without SMT and above 100 with SMT has been achieved for the improved CPU-based implementation while obtaining equivalent numerical results.

Comparison of [99mTc]HMPAO SPECT with [18F]fluoromethane PET in cerebrovascular disease.

Positron emission tomography (PET) of [18F]fluoromethane (FM) and single-photon emission tomography (SPECT) of [99mTc]hexamethylpropyleneamine oxime (HMPAO) were performed under identical conditions within 2 h in 22 patients suffering from cerebrovascular disease (8 ischemic infarction, 2 intracerebral hemorrhages, 7 transient ischemic attacks, and 5 multi-infarct syndrome). While gross pathological changes could be seen in the images of either procedure, focal abnormalities corresponding to transient ischemic deficits or to lesions in multi-infarct syndrome and areas of functional deactivation were sometimes missed on SPECT images. Overall, HMPAO SPECT images showed less contrast between high and low activity regions than the FM PET images, and differences between lesions and contralateral regions were less pronounced (6.4 vs 13.3% difference). Regional cerebral blood flow (rCBF) was calculated from FM PET studies in 14 large territorial regions and the pathological lesion, and the regional values relative to mean flow were compared to the relative HMPAO uptake in an identical set of regions defined on the SPECT images. Among individual patients, the Spearman rank-correlation coefficient between relative rCBF and HMPAO uptake varied between 0.48 and 0.89, with a mean of 0.70. While an underestimation of high flow with SPECT--which was demonstrated in a curvilinear relationship between all relative regional PET and SPECT values--could be corrected by linearization taking into account HMPAO efflux from the brain before metabolic trapping, correspondence of SPECT data with PET rCBF values was not improved since this procedure also increased the variance in high flow areas. In the cerebellum, however, a high HMPAO uptake in SPECT always overestimated CBF in relation to forebrain values; this finding might be due to high capillary density in the cerebellum. The differences observed between SPECT and PET data may be explained by technical and physical properties of the methods and by the incomplete first-pass extraction of HMPAO. Additionally, HMPAO or its metabolites may leak through a damaged blood-brain barrier (as observed in one infarct and in the surrounding of hemorrhages), impairing the contrast between lesion and normal tissue. The presented data indicate that the quantification of rCBF by HMPAO SPECT is limited.

Measurement of blood-brain hexose transport with dynamic PET: comparison of [18F]2-fluoro-2-deoxyglucose and [11C]O-methylglucose.

Blood-to-tissue transport of [18F]2-fluoro-2-deoxyglucose (FDG) and [11C]O-methylglucose (CMG) was compared by dynamic positron emission tomography in four patients with recent ischemic infarcts and in three patients with intracerebral tumors. Local blood volume, tracer transport from tissue to blood, and FDG phosphorylation rates were also determined. A regional analysis of parametric images showed a close correlation of FDG and CMG transport rate constants in pathological tissue. Transport rates of FDG and CMG showed correspondingly less asymmetric remote effects than FDG phosphorylation rates. Transport rate constants were consistently higher for FDG than for CMG in pathological and normal tissue, in accordance with the higher affinity of carrier enzymes to FDG. There was a significant correlation between fitted regional blood volume values and correspondence of average absolute values with both tracers. It is concluded that dynamic FDG PET for measurement of cerebral glucose metabolism is also useful to measure alterations of hexose transport and local blood volume in pathological tissue.

Progressive derangement of periinfarct viable tissue in ischemic stroke.

Sixteen patients were studied by multitracer positron emission tomography (PET) within 6-48 (mean of 23) h of onset of a hemispheric ischemic stroke and again 13-25 (mean of 15.6) days later. Cerebral blood flow (CBF), cerebral blood volume (CBV), cerebral metabolic rate of oxygen (CMRO2), oxygen extraction fraction (OEF), and cerebral metabolic rate of glucose (CMRglc) were measured each time by standard methods, and the sets of brain slices obtained at the two studies were matched using a three-dimensional alignment procedure. On matched brain slices, regions of interest (ROIs) for infarct and peri-infarct tissue, contralateral mirror regions, and major brain structures were outlined. In the core of infarction, blood flow and metabolism were significantly lower than in the corresponding contralateral regions at the first study, and did not change during the observation period. In the peri-infarct tissue, CMRO2 was moderately decreased at the first measurement; over time, the CMRO2 deteriorated progressively while flow did not change. When peri-infarct regions were selected on the basis of increased OEF (25 +/- 29.8% above corresponding contralateral regions) on the early scans, the CBF was significantly decreased (23 +/- 6.6%) while the CMRO2 showed only a slight difference from the mirror region. Within the observation period, the CBF improved but the CMRO2, OEF, and CMRglc deteriorated. Only in a few regions with increased OEF and slightly impaired CMRO2 was metabolism preserved close to normal values. These data from repeat PET studies in reproducibly defined tissue compartments furnish evidence of viable tissue in the border zone of ischemia up to 48 h after stroke. While this viable peri-infarct tissue exhibits some potential for effective treatment of ischemic stroke, therapeutic routines available today cannot prevent subsequent metabolic derangement and progression to necrosis. Multitracer PET studies identifying viable tissue could be of value in the development of effective treatment of ischemic stroke.

Planum temporale and Brodmann's area 22. Magnetic resonance imaging and high-resolution positron emission tomography demonstrate functional left-right asymmetry.

OBJECTIVE: To describe the registered analysis of magnetic resonance imaging and glucose metabolic data acquired with positron emission tomography to determine the relationship between structure and function of temporal lobe cortical structures between the left and right hemispheres. BACKGROUND: The dominance of the left cerebral hemisphere is associated with a preponderance of the left planum temporale. SUBJECTS AND METHODS: Fifteen subjects without signs or symptoms of a neurological disorder. Three-dimensional-registered magnetic resonance imaging and positron emission tomography with the use of fludeoxyglucose F18 and a high-resolution positron emission tomography scanner. Analysis of regional metabolic activation during single-word repetition on matched parasagittal magnetic resonance imaging and positron emission tomography. RESULTS: The planum temporale was bilaterally activated without left-right asymmetry. The metabolic increase was asymmetric within the left Brodmann's area (BA) 22. The part of the left BA 22 that was buried in the superior temporal sulcus was significantly less activated than the part of BA 22 on the surface of the superior temporal gyrus. The metabolic activation in the sulcal part of the left BA 22 had a significant inverse correlation with the anatomical predominance of the left planum temporale (r = .71, P = .003) and a significant direct correlation with the metabolic activation in the surface aspects of the right BA 22 (r = .82, P < .001). CONCLUSION: Brodmann's area 22 is a critical feature of language dominance and is also important with regard to the exchange of information between the two hemispheres.

Regional cerebral blood flow in patients with leuko-araiosis and atherosclerotic carotid artery disease.

The relation between white-matter lesions (WMLs), demonstrated with magnetic resonance imaging, and regional cerebral blood flow (CBF), measured with dynamic positron emission tomography and [18F] fluoromethane, was investigated in 20 patients with atherosclerotic disease of the internal carotid artery. There was no correlation between the extent of small patchy WMLs and hemispheric CBF, but hemispheric CBF was significantly reduced in 5 patients with multiple large or confluent lesions. Distinct focal cortical CBF reductions were observed when large WMLs (greater than 5 mm) were located directly beneath the cortex, whereas large WMLs in deeper white matter were associated with a more diffuse decrease of cortical perfusion. There was no evidence of preferential CBF reduction in vascular border zones with increasing severity of WMLs or stenosis of the internal carotid artery. The side of previous transient ischemic symptoms correlated significantly with hemispheric CBF asymmetries, but not with asymmetries of WMLs and internal carotid artery stenosis. It can be concluded from these results that the presence of small patchy WMLs shown by magnetic resonance imaging cannot be used as evidence of impaired cerebral perfusion, while large lesions indicate clinically relevant cerebrovascular disease affecting cortical blood flow.

Functional effects of striatal dysfunction in Parkinson disease.

BACKGROUND: Concepts of basal ganglia organization suggest structually and functionally segregated pathways that link putamen and caudate function to motor and cognitive performance, respectively. OBJECTIVE: To investigate whether motor and cognitive impairment in Parkinson disease is attributable to selective disturbance in nigrostriatal, dopaminergic function and regional cerebral glucose metabolism. DESIGN: Twenty patients with probable Parkinson disease underwent positron emission tomographic measurements of dopaminergic, nigrostriatal function (positron emission tomography with fluorodopa F 18), regional glucose metabolism (positron emission tomography with fludeoxyglucose F 18), memory testing, and evaluation of locomotor disability. RESULTS: Memory performance in the patient cohort strongly correlated with the individual disease duration and degree of locomotor disability (P < .05). Striatal uptake rates of fluorodopa F 18 were significantly reduced in all patients (P < .05) compared with those in normal control subjects, and putaminal rates correlated significantly with the patients' degree of locomotor disability (P < .01) but not with memory performance. In the patients with an advanced stage of disease, there was a significant correlation between reduced caudate uptake rates of fluorodopa F 18 and the patients' impairment in delayed recall performance of the memory task (P < .05) but not with the individual degree of locomotor disability. No changes were found for regional glucose metabolic rates in the patients compared with the controls. CONCLUSIONS: The present study provides evidence for the hypothesis that on the level of the striatum, motor impairment in Parkinson disease may be assigned to altered dopamine neuronal integrity in the putamen but not in the caudate, whereas memory impairment in the more advanced cases may be attributed to caudate but not putaminal dysfunction.

Epileptic negative myoclonus: An EEG-single-photon emission CT study indicating involvement of premotor cortex.

We report a combined EEG-single-photon emission CT (SPECT) study on a patient with epileptic negative myoclonus (ENM). Clinically, the ENM was characterized by brief repetitive lapses in postural tone of the right upper extremity when the arms were held outstretched, whereas no movement effect was observed during rest. Ictal EEG showed repetitive left frontal spikes with a maximum at electrodes EC1 and F1. EMG silent periods lasting from 100 to 200 ms followed the onset of the EEG transients by a latency of 20 to 40 ms. The N20 component of median nerve somatosensory evoked potentials-representing a biological marker of the location of central fissure-showed a phase reversal between electrodes P3 and C1 and thus was located considerably posterior to the spike maximum. We obtained accurate anatomic reference of cerebral blood flow changes visible on SPECT by a special coregistration technique of MRI and SPECT. SPECT performed during ENM showed a marked regional hyperperfusion in the left middle frontal gyrus and a less pronounced increase in tracer uptake in the left supramarginal gyrus. Our results suggest that ENM is generated by epileptic activity in the premotor area in the middle frontal gyrus corresponding to Brodmann's area 6.

A functional magnetic resonance imaging study of local/global processing with stimulus presentation in the peripheral visual hemifields.

When stimuli are presented in the left or right visual fields, hemispheric specialization for global and local processing in occipital areas is attenuated. Using functional magnetic resonance imaging, we investigated how this attenuation is compensated for when information must cross the corpus callosum to reach the areas specialized for global and local processing. We presented hierarchically nested letters (e.g. a large E made of smaller E's) to the right or the left visual hemifield while subjects fixated centrally. In half the trials, subjects indicated whether the global aspect and in the other half whether the local aspect of the stimulus matched a pre-specified target letter. Visual hemifield presentations showed the expected contralateral activations of occipital cortex. The main effects of locally or globally directed attention did not show any differential occipital activations, but the right anterior cingulate cortex was activated differentially during local processing. Region-of-interest-based analyses showed increased neural activity in left posterior occipital cortex during local processing when stimuli were presented in the left hemifield. During global processing with stimulus presentation to the right hemifield, the right posterior occipital cortex was activated. Activation of right anterior cingulate cortex during local processing is likely to reflect the suppression of global processing precedence in order to select correctly the local stimulus level. The activations in left (local) and right (global) occipital areas are likely to reflect the top-down augmentation of stimulus information that has been degraded by callosal crossing in order to access the hemisphere specialized for local or global processing.

Imaging of cerebral blood flow with technetium-99m-HMPAO and technetium-99m-ECD: a comparison.

UNLABELLED: Because 99mTc-HMPAO and 99mTc-ECD are both used for SPECT imaging of cerebral blood flow, the question arises whether there are any differences in their respective regional cerebral distribution. For that purpose, visual and semiquantitative comparisons between 99mTc-HMPAO and 99mTc-ECD studies were performed. METHODS: Seventeen patients (4 women; 13 men; age 45-89 yr; mean age 71 yr) with various neurological diseases, except acute/subacute stroke, were investigated twice with 99mTc-HMPAO and 99mTc-ECD using a triple-headed rotating SPECT camera. After image reorientation, the two studies were evaluated visually. Seventy regions of interest (ROIs) were drawn manually and the same set of ROIs was applied in both studies. Regional indices (RI) normalized to individual brain values were calculated and first compared between two random patient groups. Second, for all patients, RI for 70 and later for 27 regions (gained after summing values of corresponding regions in different brain slices) were compared by using a paired Student's t-test applying Bonferroni's correction. RESULTS: Visual evaluation demonstrated relatively high 99mTc-ECD uptake in occipital and comparatively low uptake in mediotemporal regions. Calculation of RI revealed significantly higher values in the right cerebellum, brainstem, mediotemporal regions, right basal ganglia and the thalamus in the 99mTc-HMPAO SPECT studies and higher values in the occipital, supratemporal/inferior parietal and parietal cortex in the 99mTc-ECD SPECT studies, respectively. CONCLUSION: Significant differences in regional tracer distribution between 99mTc-HMPAO and 99mTc-ECD could be detected, probably caused by different tracer kinetics. The results indicate that direct comparisons of studies performed with 99mTc-HMPAO and 99mTc-ECD are not possible and the use of either tracer can be favorable in different clinical questions.

Presurgical visualization of primary breast carcinoma with PET emission and transmission imaging.

UNLABELLED: The aim of this study was to investigate a technique that visualizes findings from PET images in a context useful for surgery. METHODS: Simultaneously acquired PET emission and transmission scans were used. By applying a multipurpose imaging, registration and rendering tool (MPM), displays of orthogonal and volume-rendered views or any combination thereof were obtained. The PET emission and transmission scans were acquired under routine conditions. The final user-customized display (with a combination of orthogonal cuts and rendered views) was processed in 10 min or less on commercially available hardware. Distinct features of the body shape were clearly visible on the volume-rendered transmission views. Hot spots, e.g., in primary breast cancer, from the emission scans could be easily assessed in their localization relative to the body outline. CONCLUSION: Rendering of the main signatures in a single comprehensive display makes this method potentially valuable for simple presurgical workup and therapeutic management of breast cancer.

F-Dopa as an amino acid tracer to detect brain tumors.

A 57-yr-old woman suffering from light movement disorder of the left arm and hand was referred for 18F-Dopa PET. The PET study not only proved asymmetrically reduced dopamine uptake in the putamen (influx constant Ki right 0.0064/min, left 0.0086) but also revealed pathologically increased 18F-Dopa accumulation in the right frontal lobe. Further PET examinations demonstrated increased 11C-methionine uptake and low glucose metabolism in this right frontal region. MRI and 1H-MRSI showed a heterogeneous lesion with reduced N-acetyl-aspartate and increased choline and lactate, suggesting a mixed, low-grade glioma. In 15O-water studies, during intentional movements of one hand the respective motor areas were identified, indicating asymmetries due to the mass occupying lesion. The tumor could be removed in open surgery, thus sparing the motor areas; a mild postoperative motor deficit resolved to the presurgical state. Histology confirmed the diagnosis of a grade 2 oligo-astrocytoma. This case impressively demonstrates that 18F-Dopa can be used as an amino acid tracer for brain tumor detection in addition to its established application to assess aromatic acid decarboxylase activity.

High-resolution PET in cats: application of a clinical camera to experimental studies.

UNLABELLED: A commercial high-resolution scanner designed for clinical PET studies was tested for its applicability to investigate cerebral metabolism and blood flow in cats. METHODS: Cerebral blood flow, CMRO2, CBV and CMRglc were determined repeatedly using 15O steady-state oxygen methods and 18F-fluorodeoxyglucose (FDG). Metabolic and blood flow images of 14 contiguous 3-mm PET slices were compared to histological sections in four control animals. In another six cats, hemodynamic and metabolic changes were followed by serial multi-tracer PET for 24 hr after permanent occlusion of the left middle cerebral artery (MCA). Pattern and extent of changes of the physiological variables were related to the final infarct verified in matched histological sections. RESULTS: At spatial resolutions (FWHM) of 3.6 mm in transaxial planes and 4.0 mm axially, details of the gross anatomy of the cat brain were distinguished best in the FDG images. Cerebral blood flow, CMRO2 and CMRglc values measured in the cortex, white matter and basal ganglia were in the range of common autoradiographic results. Immediately after MCA occlusion, there was widespread decrease in blood flow, but metabolism was preserved at values, which suggest viable tissue. With time, the areas of increased oxygen extraction fraction (OEF) moved from the center to the periphery of the MCA territory. CONCLUSION: High-resolution PET can be used for repeat, quantitative imaging of blood flow and metabolism in small animals such as the cat. After MCA occlusion, the changes in blood flow and metabolism can be followed over time and can be related to the final morphological lesion.

An interactive technique for three-dimensional image registration: validation for PET, SPECT, MRI and CT brain studies.

UNLABELLED: A multipurpose three-dimensional registration technique was validated with PET, SPECT, CT and MRI scans, which had been obtained under normal clinical conditions. In contrast to fully automated procedures, this coregistration method is highly interactive, which has the advantage that it does not impose rigid restrictions by data type and by alterations in normal anatomy or brain function resulting from disease. METHODS: Basically, a computer program provides a variety of tools to examine the accuracy of coregistration visually and to specify necessary translations and rotations in all three dimensions. Tools and criteria to accept coregistration were applied according to a standardized protocol. Reproducibility was assessed with five independent users on nine pairs of image sets. In two pairs of these image sets, coregistration was repeated three times by each user. RESULTS: Depending on the resolution of the images involved, the reproducibility of translation distances ranged from 0.32 to 2.22 mm (s.d.) and of rotation angles from 0.32 to 1.70 degrees. It was always much smaller than the point-spread full-width half maximum of the device with the lower resolution. The accuracy of coregistration was examined using two arbitrarily misplaced image sets. Interindividual and intraindividual variance were similar, which suggested that the influence of subjectivity was not significant. Average displacements after coregistration were 0.43 and 0.29 mm or less for PET and MRI data, respectively, which indicated the absence of a systematic bias. CONCLUSION: The results indicate the high reproducibility and accuracy of this three-dimensional coregistration technique, which is comparable or superior to those of automated techniques and methods based on external artificial landmarks.

Preictal SPECT in temporal lobe epilepsy: regional cerebral blood flow is increased prior to electroencephalography-seizure onset.

UNLABELLED: Peri-ictal SPECT provides unique information on the dynamic changes in regional cerebral blood flow (rCBF) that occur during seizure evolution and, thus, could be useful in clarifying the poorly understood interplay of the interictal and ictal states in human focal epilepsy. The regional hyperperfusion observed on ictal SPECT is generally believed to be a consequence of electrical seizure activity. However, recent studies using invasive long-term cortical CBF monitoring have demonstrated that rCBF changes occur up to 20 min prior to ictal electroencephalography (EEG) onset. Because of apparent technical difficulties, no preictal SPECT studies have been reported so far. Therefore, we present our results on two patients with temporal lobe epilepsy in whom preictal SPECT scans were performed fortuitously under continuous video-EEG monitoring control. METHODS: Technetium-99m-hexamethyl propyleneamine oxime was injected 11 min (Patient 1) and 12 min (Patient 2) before clinical and EEG seizure onset, as documented from simultaneous video-EEG monitoring in two patients with temporal lobe epilepsy. We obtained accurate anatomical reference of CBF changes visible on SPECT by a special coregistration technique of MRI and SPECT. RESULTS: Whereas interictal SPECT showed a hypoperfusion of the temporal lobe ipsilateral to the seizure focus, on preictal SPECT, a significant increase in rCBF in the epileptic temporal lobe could be observed. These rCBF changes were not accompanied by any significant changes of the ongoing EEG. CONCLUSION: Our study provides evidence that rCBF is increased in the epileptic temporal lobe several minutes before EEG seizure onset. Thus, rCBF changes observed on peri-ictal SPECT scan cannot be considered a mere consequence of EEG seizure activity but may rather reflect a change in neuronal activity precipitating the transition from the interictal to the ictal state.

A positron emission tomography study of oculomotor imagery.

Focal increases of regional cerebral blood flow (rCBF) were measured by positron emission tomography in order to study the anatomo-functional analogies between imagined and executed saccadic eye movements. Oculomotor imagery was performed in the absence of overt eye movements. Compared with a control state the two conditions were associated with normalized rCBF increases in the median cingulate gyrus, and the supplementary and frontal eye fields of both hemispheres. Therefore in the human brain execution and mental imagery of eye movement appear to be functionally linked and mediated by a common network of frontal structures.