Hans-W. Müller-Gärtner

Expectation maximization reconstruction of positron emission tomography images using anatomical magnetic resonance information

Using statistical methods the reconstruction of positron emission tomography (PET) images can be improved by high-resolution anatomical information obtained from magnetic resonance (MR) images. The authors implemented two approaches that utilize MR data for PET reconstruction. The anatomical MR information is modeled as a priori distribution of the PET image and combined with the distribution of the measured PET data to generate the a posteriori function from which the expectation maximization (EM)-type algorithm with a maximum a posteriori (MAP) estimator is derived. One algorithm (Markov-GEM) uses a Gibbs function to model interactions between neighboring pixels within the anatomical regions. The other (Gauss-EM) applies a Gauss function with the same mean for all pixels in a given anatomical region. A basic assumption of these methods is that the radioactivity is homogeneously distributed inside anatomical regions. Simulated and phantom data are investigated under the following aspects: count density, object size, missing anatomical information, and misregistration of the anatomical information. Compared with the maximum likelihood-expectation maximization (ML-EM) algorithm the results of both algorithms show a large reduction of noise with a better delineation of borders. Of the two algorithms tested, the Gauss-EM method is superior in noise reduction (up to 50%). Regarding incorrect a priori information the Gauss-EM algorithm is very sensitive, whereas the Markov-GEM algorithm proved to be stable with a small change of recovery coefficients between 0.5 and 3%.

Quantitation of Regional Cerebral Blood Flow with 15O-Butanol and Positron Emission Tomography in Humans

We describe the implementation and validation of a combined dynamic–autoradiographic approach for measuring the regional cerebral blood flow (rCBF) with 15O-butanol. From arterial blood data sampled at a rate of 1 s and list mode data of the cerebral radioactivity accumulated over 100 s, the time shift between blood and tissue curves, the dispersion constant DC, the partition coefficient p, and the CBF were estimated by least squares fitting. Using the fit results, a pixel-by-pixel parametrization of rCBF was computed for a single 40-s (autoradiographic) 15O-butanol uptake image. The mean global CBF found in 27 healthy subjects was 49 ± 8 ml 100 g−1 min−1. Gray and white matter rCBF were 83 ± 20 and 16 ± 3 ml 100 g−1 min−1, respectively, with a corresponding partition coefficient p of 0.77 ± 0.18 and 0.77 ± 0.29 ml/g in both compartments. The quantitative images resulted in a significantly higher gray matter rCBF than the autoradiographic images.

4.5 tesla magnetic field reduces range of high-energy positrons-potential implications for positron emission tomography

We have theoretically and experimentally investigated the extent to which homogeneous magnetic fields up to 7 Tesla reduce the spatial distance positrons travel before annihilation (positron range). Computer simulations of a noncoincident detector design using a Monte Carlo algorithm calculated the positron range as a function of positron energy and magnetic field strength. The simulation predicted improvements in resolution, defined as full-width at half-maximum (FWBM) of the line-spread function (LSP) for a magnetic field strength up to 7 Tesla: negligible for F-18, from 3.35 mm to 2.73 mm for Ga-68 and from 3.66 mm to 2.68 mm for Rb-82. Also a substantial noise suppression was observed, described by the full-width at tenth-maximum (FWTM) for higher positron energies. The experimental approach confirmed an improvement in resolution for Ga-68 from 3.54 mm at 0 Tesla to 2.99 mm FWHM at 4.5 Tesla and practically no improvement for F-18 (2.97 mm at 0 Tesla and 2.95 mm at 4.5 Tesla). It is concluded that the simulation model is appropriate and that a homogeneous static magnetic field of 4.5 Tesla reduces the range of high-energy positrons to an extent that may improve spatial resolution in positron emission tomography.

Imaging dopamine D4 receptors in the living primate brain: A positron emission tomography study using the novel D1/ D4 antagonist [11C]SDZ GLC 756

The dopamine D4 receptor has lately attracted interest since it has been hypothesized to be involved in the pathogenesis and pharmacotherapy of neuropsychiatric diseases. The present study provides first in vivo evidence of dopamine D4 receptors in primate brain using a [11C]benzo[g]quinoline, the novel radioligand [11C]SDZ GLC 756 ([11C]GLC: in vitro dissociation constants at human receptor clones [nM]: 1.10 at D1; 0.40 at D2; 25 at D3; 0.18 at D4.2; 6.03 at D5). Dynamic positron emission tomography scans were performed on healthy baboons (Papio hamadryas, n = 3). Specific receptor binding (SB) was calculated for striatum and neocortex (frontal, temporal, parietal, and occipital) based on the differences between the regional and the cerebellar concentration of [11C]. Blockade of D1 and D5 receptors by SCH23390 (1.7 μmol/ kg) diminished SB in the striatum by 55 ± 4% (mean ± standard deviation, P < 0.05) and in the frontal cortex by 13 ± 8% (P < 0.05) when compared to SB in the unblocked state (SBD1–D5). In the presence of the dopamine antagonists SCH23390 (1.7 μmol/ kg) and raclopride (5.7 μmol/ kg)—which mask the D1, D2, D3, and D5 subtypes—SB of [11C]GLC to D4 receptors (SBD4) was demonstrated in the striatum and all cortical regions of interest. In the striatum, the ratio of SBD4/ SBD1–D5 was 0.13 ± 0.07. In the neocortex, SBD4/ SBD1–D5 was notably higher (0.77 ±0.29; mean of all cortical regions of interest). The widespread distribution of dopamine D4 receptors suggests a basic functional role of this receptor subtype in the modulation of cortical and subcortical neuronal activity. Synapse 30:341–350, 1998.

Imaging techniques in the analysis of brain function and behaviour

Techniques such as positron-emission tomography, single-photon-emission computed tomography, functional magnetic-resonance imaging and magnetoencephalography permit the observation of biological processes in the brain in a noninvasive manner. They have yielded new insights into the biological interrelations of sensory, motor and cognitive functions, as well as into brain diseases. Combined use of these techniques may provide more information than just the sum of its constituents, and this may narrow the gap between the biological data provided by these techniques and the mental models described by clinicians, mathematicians, psychologists and philosophers.

Calculation of residence times and radiation doses using the standard PC software Excel

Abstract. We developed a program which aims to facilitate the calculation of radiation doses to single organs and the whole body. IMEDOSE uses Excel to include calculations, graphical displays, and interactions with the user in a single general-purpose PC software tool. To start the procedure the input data are copied into a spreadsheet. They must represent percentage uptake values of several organs derived from measurements in animals or humans. To extrapolate these data up to seven half-lives of the radionuclide, fitting to one or two exponentional functions is included and can be checked by the user. By means of the approximate time-activity information the cumulated activity or residence times are calculated. Finally these data are combined with the absorbed fraction doses (S-values) given by MIRD pamphlet No. 11 to yield radiation doses, the effective dose equivalent and the effective dose. These results are presented in a final table. Interactions are realized with push-buttons and drop-down menus. Calculations use the Visual Basic tool of Excel. In order to test our program, biodistribution data of fluorine-18 fluorodeoxyglucose were taken from the literature (Meija et al., J Nucl Med 1991; 32:699–706). For a 70-kg adult the resulting radiation doses of all target organs listed in MIRD 11 were different from the ICRP 53 values by 1%±18% on the average. When the residence times were introduced into MIRDOSE3 (Stabin, J Nucl Med 1996; 37:538–546) the mean difference between our results and those of MIRDOSE3 was –3%±6%. Both outcomes indicate the validity of the present approach.

Influence of acetylcholine on binding of 4-[125I]iododexetimide to muscarinic brain receptors.

The distribution of nicotinic and muscarinic cholinergic receptors in the human brain in vivo has been successfully characterized using radiolabeled tracers and emission tomography. The effect of acetylcholine release into the synaptic cleft on receptor binding of these tracers has not yet been investigated. The present study examined the influence of acetylcholine on binding of 4-[125I]iododexetimide to muscarinic cholinergic receptors of porcine brain synaptosomes in vitro. 4-Iododexetimide is a subtype-unspecific muscarinic receptor antagonist with high affinity. Acetylcholine competed with 4-[125I]iododexetimide in a dose-dependent manner. A concentration of 500 microM acetylcholine inhibited 50% of total specific 4-[125I]iododexetimide binding to synaptosomes when both substances were given simultaneously. An 800 microM acetylcholine solution reduced total specific 4-[125I]iododexetimide binding by about 35%, when acetylcholine was given 60 min after incubation of synaptosomes with 4-[125I]iododexetimide. Variations in the synaptic acetylcholine concentration might influence muscarinic cholinergic receptor imaging in vivo using 4-[123I]iododexetimide. Conversely, 4-[123I]iododexetimide might be an appropriate molecule to investigate alterations of acetylcholine release into the synaptic cleft in vivo using single photon emission computed tomography.

Comparison of 123I-α-Methyltyrosine SPECT and 11C-L-Methionine PET in Patients with Brain Tumors

PET studies using 11C-L-methionine (MET) have been shown to yield useful information in patients with brain tumors. The aim of this study was to investigate, whether SPECT studies using 123I-α-methyltyrosine (IMT) can produce similar results.