Globally convergent algorithms for maximum a posteriori transmission tomography

Statistical image reconstruction from limited projection data with intensity priors

The radiation dose generated from x-ray computed tomography (CT) scans and its responsibility for increasing the risk of malignancy became a major concern in the medical imaging community. Accordingly, investigating possible approaches for image reconstruction from low-dose imaging protocols, which minimize the patient radiation exposure without affecting image quality, has become an issue of interest. Statistical reconstruction (SR) methods are known to achieve a superior image quality compared with conventional analytical methods. Effective physical noise modeling and possibilities of incorporating priors in the image reconstruction problem are the main advantages of the SR methods. Nevertheless, the high computation cost limits its wide use in clinical scanners. This paper presents a framework for SR in x-ray CT when the angular sampling rate of the projection data is low. The proposed framework is based on the fact that, in many CT imaging applications, some physical and anatomical structures and the corresponding attenuation information of the scanned object can be a priori known. Therefore, the x-ray attenuation distribution in some regions of the object can be expected prior to the reconstruction. Under this assumption, the proposed method is developed by incorporating this prior information into the image reconstruction objective function to suppress streak artifacts. We limit the prior information to only a set of intensity values that represent the average intensity of the normal and expected homogeneous regions within the scanned object. This prior information can be easily computed in several x-ray CT applications. Considering the theory of compressed sensing, the objective function is formulated using the ℓ(1) norm distance between the reconstructed image and the available intensity priors. Experimental comparative studies applied to simulated data and real data are used to evaluate the proposed method. The comparison indicates a significant improvement in image quality when the proposed method is used.

Joint L1 and total variation regularization for fluorescence molecular tomography

Fluorescence molecular tomography (FMT) is an imaging modality that exploits the specificity of fluorescent biomarkers to enable 3D visualization of molecular targets and pathways in vivo in small animals. Owing to the high degree of absorption and scattering of light through tissue, the FMT inverse problem is inherently ill-conditioned making image reconstruction highly susceptible to the effects of noise and numerical errors. Appropriate priors or penalties are needed to facilitate reconstruction and to restrict the search space to a specific solution set. Typically, fluorescent probes are locally concentrated within specific areas of interest (e.g., inside tumors). The commonly used L(2) norm penalty generates the minimum energy solution, which tends to be spread out in space. Instead, we present here an approach involving a combination of the L(1) and total variation norm penalties, the former to suppress spurious background signals and enforce sparsity and the latter to preserve local smoothness and piecewise constancy in the reconstructed images. We have developed a surrogate-based optimization method for minimizing the joint penalties. The method was validated using both simulated and experimental data obtained from a mouse-shaped phantom mimicking tissue optical properties and containing two embedded fluorescent sources. Fluorescence data were collected using a 3D FMT setup that uses an EMCCD camera for image acquisition and a conical mirror for full-surface viewing. A range of performance metrics was utilized to evaluate our simulation results and to compare our method with the L(1), L(2) and total variation norm penalty-based approaches. The experimental results were assessed using the Dice similarity coefficients computed after co-registration with a CT image of the phantom.

Iterative reconstruction methods in X-ray CT

Iterative reconstruction (IR) methods have recently re-emerged in transmission x-ray computed tomography (CT). They were successfully used in the early years of CT, but given up when the amount of measured data increased because of the higher computational demands of IR compared to analytical methods. The availability of large computational capacities in normal workstations and the ongoing efforts towards lower doses in CT have changed the situation; IR has become a hot topic for all major vendors of clinical CT systems in the past 5 years. This review strives to provide information on IR methods and aims at interested physicists and physicians already active in the field of CT. We give an overview on the terminology used and an introduction to the most important algorithmic concepts including references for further reading. As a practical example, details on a model-based iterative reconstruction algorithm implemented on a modern graphics adapter (GPU) are presented, followed by application examples for several dedicated CT scanners in order to demonstrate the performance and potential of iterative reconstruction methods. Finally, some general thoughts regarding the advantages and disadvantages of IR methods as well as open points for research in this field are discussed.

Two-view and single-view tomosynthesis versus full-field digital mammography: high-resolution X-ray imaging observer study

PURPOSE

To compare the diagnostic accuracy of two-dimensional (2D) full-field digital mammography with that of two-view (mediolateral and craniocaudal) and single-view (mediolateral oblique) tomosynthesis in an observer study involving two institutions.

MATERIALS AND METHODS

Ethical committee approval was obtained. All participating women gave informed consent. Two hundred twenty women (mean age, 56.3; range, 40-80 years) with breast density of 2-4 according to American College of Radiology criteria were recruited between November 2008 and September 2009 and underwent standard treatment plus tomosynthesis with a prototype photon-counting machine. After exclusion criteria were met, this resulted in a final test set of 130 women. Ten accredited readers classified the 130 cases (40 cancers, 24 benign lesions, and 66 normal images) using 2D mammography and two-view tomosynthesis. Another 10 readers reviewed the same cases using 2D mammography but single-view tomosynthesis. The multireader, multicase receiver operating characteristic (ROC) method was applied. The significance of the observed difference in accuracy between 2D mammography and tomosynthesis was calculated.

RESULTS

For diagnostic accuracy, 2D mammography performed significantly worse than two-view tomosynthesis (average area under ROC curve [AUC] = 0.772 for 2D, AUC = 0.851 for tomosynthesis, P = .021). Significant differences were found for both masses and microcalcification (P = .037 and .049). The difference in AUC between the two modalities of -0.110 was significant (P = .03) only for the five readers with the least experience (<10 years of reading); with AUC of -0.047 for the five readers with 10 years or more experience (P = .25). No significant difference (P = .79) in reader performance was seen when 2D mammography (average AUC = 0.774) was compared with single-view tomosynthesis (average AUC = 0.775).

CONCLUSION

Two-view tomosynthesis outperforms 2D mammography but only for readers with the least experience. The benefits were seen for both masses and microcalcification. No differences in classification accuracy was seen between and 2D mammography and single-view tomosynthesis.

Gaussian-mixture-model-based spatial neighborhood relationships for pixel labeling problem

In this paper, we present a new algorithm for pixel labeling and image segmentation based on the standard Gaussian mixture model (GMM). Unlike the standard GMM where pixels themselves are considered independent of each other and the spatial relationship between neighboring pixels is not taken into account, the proposed method incorporates this spatial relationship into the standard GMM. Moreover, the proposed model requires fewer parameters compared with the models based on Markov random fields. In order to estimate model parameters from observations, instead of utilizing an expectation-maximization algorithm, we employ gradient method to minimize a higher bound on the data negative log-likelihood. The performance of the proposed model is compared with methods based on both standard GMM and Markov random fields, demonstrating the robustness, accuracy, and effectiveness of our method.

Recent advances in iterative reconstruction for clinical SPECT/PET and CT

Statistical iterative reconstruction is now widely used in clinical practice and has contributed to significant improvement in image quality in recent years. Although primarily used for reconstruction in emission tomography (both single photon emission computed tomography (SPECT) and positron emission tomography (PET)) there is increasing interest in also applying similar algorithms to x-ray computed tomography (CT). There is increasing complexity in the factors that are included in the reconstruction, a demonstration of the versatility of the approach. Research continues with exploration of methods for further improving reconstruction quality with effective correction for various sources of artefact.

Noise-resolution tradeoffs in x-ray CT imaging: a comparison of penalized alternating minimization and filtered backprojection algorithms

PURPOSE

In comparison with conventional filtered backprojection (FBP) algorithms for x-ray computed tomography (CT) image reconstruction, statistical algorithms directly incorporate the random nature of the data and do not assume CT data are linear, noiseless functions of the attenuation line integral. Thus, it has been hypothesized that statistical image reconstruction may support a more favorable tradeoff than FBP between image noise and spatial resolution in dose-limited applications. The purpose of this study is to evaluate the noise-resolution tradeoff for the alternating minimization (AM) algorithm regularized using a nonquadratic penalty function.

METHODS

Idealized monoenergetic CT projection data with Poisson noise were simulated for two phantoms with inserts of varying contrast (7%-238%) and distance from the field-of-view (FOV) center (2-6.5 cm). Images were reconstructed for the simulated projection data by the FBP algorithm and two penalty function parameter values of the penalized AM algorithm. Each algorithm was run with a range of smoothing strengths to allow quantification of the noise-resolution tradeoff curve. Image noise is quantified as the standard deviation in the water background around each contrast insert. Modulation transfer functions (MTFs) were calculated from six-parameter model fits to oversampled edge-spread functions defined by the circular contrast-insert edges as a metric of local resolution. The integral of the MTF up to 0.5 1p/mm was adopted as a single-parameter measure of local spatial resolution.

RESULTS

The penalized AM algorithm noise-resolution tradeoff curve was always more favorable than that of the FBP algorithm. While resolution and noise are found to vary as a function of distance from the FOV center differently for the two algorithms, the ratio of noises when matching the resolution metric is relatively uniform over the image. The ratio of AM-to-FBP image variances, a predictor of dose-reduction potential, was strongly dependent on the shape of the AM's nonquadratic penalty function and was also strongly influenced by the contrast of the insert for which resolution is quantified. Dose-reduction potential, reported here as the fraction (%) of FBP dose necessary for AM to reconstruct an image with comparable noise and resolution, for one penalty parameter value of the AM algorithm was found to vary from 70% to 50% for low-contrast and high-contrast structures, respectively, and from 70% to 10% for the second AM penalty parameter value. However, the second penalty, AM-700, was found to suffer from poor low-contrast resolution when matching the high-contrast resolution metric with FBP.

CONCLUSIONS

The results of this simulation study imply that penalized AM has the potential to reconstruct images with similar noise and resolution using a fraction (10%-70%) of the FBP dose. However, this dose-reduction potential depends strongly on the AM penalty parameter and the contrast magnitude of the structures of interest. In addition, the authors' results imply that the advantage of AM can be maximized by optimizing the nonquadratic penalty function to the specific imaging task of interest. Future work will extend the methods used here to quantify noise and resolution in images reconstructed from real CT data.

Selective-diffusion regularization for enhancement of microcalcifications in digital breast tomosynthesis reconstruction

PURPOSE

Digital breast tomosynthesis (DBT) has been shown to improve mass detection. Detection of microcalcifications is more challenging because of the large breast volume to be searched for subtle signals. The simultaneous algebraic reconstruction technique (SART) was found to provide good image quality for DBT, but the image noise is amplified with an increasing number of iterations. In this study, the authors developed a selective-diffusion (SD) method for noise regularization with SART to improve the contrast-to-noise ratio (CNR) of microcalcifications in the DBT slices for human or machine detection.

METHODS

The SD method regularizes SART reconstruction during updating with each projection view. Potential microcalcifications are differentiated from the noisy background by estimating the local gradient information. Different degrees of regularization are applied to the signal or noise classes, such that the microcalcifications will be enhanced while the noise is suppressed. The new SD method was compared to several current methods, including the quadratic Laplacian (QL) method, the total variation (TV) method, and the nonconvex total p-variation (TpV) method for noise regularization with SART. A GE GEN2 prototype DBT system with a stationary digital detector was used for the acquisition of DBT scans at 21 angles in 3 degrees increments over a +/-30 degrees range. The reconstruction image quality without regularization and that with the different regularization methods were compared using the DBT scans of an American College of Radiology phantom and a human subject. The CNR and the full width at half maximum (FWHM) of the line profiles of microcalcifications within the in-focus DBT slices were used as image quality measures.

RESULTS

For the comparison of large microcalcifications in the DBT data of the subject, the SD method resulted in comparable CNR to the nonconvex TpV method. Both of them performed better than the other two methods. For subtle microcalcifications, the SD method was superior to other methods in terms of CNR. In both the subject and phantom DBT data, for large microcalcifications, the FWHM of the SD method was comparable to that without regularization, which was wider than that of the TV type methods. For subtle microcalcifications, the SD method had comparable FWHM values to the TV type methods. All three regularization methods were superior to the QL method in terms of FWHM.

CONCLUSIONS

The SART regularized by the selective-diffusion method enhanced the CNR and preserved the sharpness of microcalcifications. In comparison with three existing regularization methods, the selective-diffusion regularization was superior to the other methods for subtle microcalcifications.

New iterative cone beam CT reconstruction software: parameter optimisation and convergence study

Cone beam computed tomography (CBCT) provides a volumetric image reconstruction from tomographic projection data. Image quality is the main concern for reconstruction in comparison to conventional CT. The reconstruction algorithm used is clearly important and should be carefully designed, developed and investigated before it can be applied clinically. The Multi-Instrument Data Analysis System (MIDAS) tomography software originally designed for geophysical applications has been modified to CBCT image reconstruction. In CBCT reconstruction algorithms, iterative methods offer the potential to generate high quality images and would be an advantage especially for down-sampling projection data. In this paper, studies of the CBCT iterative algorithms implemented in MIDAS are presented. Stability, convergence rate, quality of reconstructed image and edge recovery are suggested as the main criteria for monitoring reconstructive performance. Accordingly, the selection of relaxation parameter and number of iterations are studied in detail. Results are presented, where images are reconstructed from full and down-sampled cone beam CT projection data using iterative algorithms. Various iterative algorithms have been implemented and the best selection of the iteration number and relaxation parameters are investigated for ART. Optimal parameters are chosen where the errors in projected data as well as image errors are minimal.

Simultaneous segmentation and reconstruction: a level set method approach for limited view computed tomography

PURPOSE

An iterative tomographic reconstruction algorithm that simultaneously segments and reconstructs the reconstruction domain is proposed and applied to tomographic reconstructions from a sparse number of projection images.

METHODS

The proposed algorithm uses a two-phase level set method segmentation in conjunction with an iterative tomographic reconstruction to achieve simultaneous segmentation and reconstruction. The simultaneous segmentation and reconstruction is achieved by alternating between level set function evolutions and per-region intensity value updates. To deal with the limited number of projections, a priori information about the reconstruction is enforced via penalized likelihood function. Specifically, smooth function within each region (piecewise smooth function) and bounded function intensity values for each region are assumed. Such a priori information is formulated into a quadratic objective function with linear bound constraints. The level set function evolutions are achieved by artificially time evolving the level set function in the negative gradient direction; the intensity value updates are achieved by using the gradient projection conjugate gradient algorithm.

RESULTS

The proposed simultaneous segmentation and reconstruction results were compared to "conventional" iterative reconstruction (with no segmentation), iterative reconstruction followed by segmentation, and filtered backprojection. Improvements of 6%-13% in the normalized root mean square error were observed when the proposed algorithm was applied to simulated projections of a numerical phantom and to real fan-beam projections of the Catphan phantom, both of which did not satisfy the a priori assumptions.

CONCLUSIONS

The proposed simultaneous segmentation and reconstruction resulted in improved reconstruction image quality. The algorithm correctly segments the reconstruction space into regions, preserves sharp edges between different regions, and smoothes the noise within each region. The proposed algorithm framework has the flexibility to be adapted to different a priori constraints while maintaining the benefits achieved by the simultaneous segmentation and reconstruction.

Evaluation of scatter effects on image quality for breast tomosynthesis

Digital breast tomosynthesis uses a limited number (typically 10-20) of low-dose x-ray projections to produce a pseudo-three-dimensional volume tomographic reconstruction of the breast. The purpose of this investigation was to characterize and evaluate the effect of scattered radiation on the image quality for breast tomosynthesis. In a simulation, scatter point spread functions generated by a Monte Carlo simulation method were convolved over the breast projection to estimate the distribution of scatter for each angle of tomosynthesis projection. The results demonstrate that in the absence of scatter reduction techniques, images will be affected by cupping artifacts, and there will be reduced accuracy of attenuation values inferred from the reconstructed images. The effect of x-ray scatter on the contrast, noise, and lesion signal-difference-to-noise ratio (SDNR) in tomosynthesis reconstruction was measured as a function of the tumor size. When a with-scatter reconstruction was compared to one without scatter for a 5 cm compressed breast, the following results were observed. The contrast in the reconstructed central slice image of a tumorlike mass (14 mm in diameter) was reduced by 30%, the voxel value (inferred attenuation coefficient) was reduced by 28%, and the SDNR fell by 60%. The authors have quantified the degree to which scatter degrades the image quality over a wide range of parameters relevant to breast tomosynthesis, including x-ray beam energy, breast thickness, breast diameter, and breast composition. They also demonstrate, though, that even without a scatter rejection device, the contrast and SDNR in the reconstructed tomosynthesis slice are higher than those of conventional mammographic projection images acquired with a grid at an equivalent total exposure.

Radiation dose reduction in computed tomography: techniques and future perspective

Despite universal consensus that computed tomography (CT) overwhelmingly benefits patients when used for appropriate indications, concerns have been raised regarding the potential risk of cancer induction from CT due to the exponentially increased use of CT in medicine. Keeping radiation dose as low as reasonably achievable, consistent with the diagnostic task, remains the most important strategy for decreasing this potential risk. This article summarizes the general technical strategies that are commonly used for radiation dose management in CT. Dose-management strategies for pediatric CT, cardiac CT, dual-energy CT, CT perfusion and interventional CT are specifically discussed, and future perspectives on CT dose reduction are presented.

Performance comparison between total variation (TV)-based compressed sensing and statistical iterative reconstruction algorithms

Of all available reconstruction methods, statistical iterative reconstruction algorithms appear particularly promising since they enable accurate physical noise modeling. The newly developed compressive sampling/compressed sensing (CS) algorithm has shown the potential to accurately reconstruct images from highly undersampled data. The CS algorithm can be implemented in the statistical reconstruction framework as well. In this study, we compared the performance of two standard statistical reconstruction algorithms (penalized weighted least squares and q-GGMRF) to the CS algorithm. In assessing the image quality using these iterative reconstructions, it is critical to utilize realistic background anatomy as the reconstruction results are object dependent. A cadaver head was scanned on a Varian Trilogy system at different dose levels. Several figures of merit including the relative root mean square error and a quality factor which accounts for the noise performance and the spatial resolution were introduced to objectively evaluate reconstruction performance. A comparison is presented between the three algorithms for a constant undersampling factor comparing different algorithms at several dose levels. To facilitate this comparison, the original CS method was formulated in the framework of the statistical image reconstruction algorithms. Important conclusions of the measurements from our studies are that (1) for realistic neuro-anatomy, over 100 projections are required to avoid streak artifacts in the reconstructed images even with CS reconstruction, (2) regardless of the algorithm employed, it is beneficial to distribute the total dose to more views as long as each view remains quantum noise limited and (3) the total variation-based CS method is not appropriate for very low dose levels because while it can mitigate streaking artifacts, the images exhibit patchy behavior, which is potentially harmful for medical diagnosis.

PET transmission tomography using a novel nonlocal MRF prior

In positron emission tomography, transmission scans can be performed to estimate attenuation correction factors (ACFs) which are in turn used to correct the emission scans. And such an attenuation correction is crucial for quantitatively accurate PET reconstructions. The prior model used in this work was based on our assumption that the attenuation values vary smoothly, with occasional discontinuities at anatomical borders. And on the other hand, long acquisition or scan times, although alleviating the noise effect of the count-limited scans, are blamed for patient uncomfortableness and movements. So, transmission tomography often suffers from the noise effect because of the short scan time. Thus reconstruction which is capable of overcoming the noise effect is highly needed. In this article, we apply the nonlocal prior Bayesian reconstruction method in PET transmission tomography. Resulting experimentations validate that the reconstructions using the nonlocal prior can reconstruct better transmission images and overcome noise effect even when the scan time is relatively short.

Electronic noise modeling in statistical iterative reconstruction

We consider electronic noise modeling in tomographic image reconstruction when the measured signal is the sum of a Gaussian distributed electronic noise component and another random variable whose log-likelihood function satisfies a certain linearity condition. Examples of such likelihood functions include the Poisson distribution and an exponential dispersion (ED) model that can approximate the signal statistics in integration mode X-ray detectors. We formulate the image reconstruction problem as a maximum-likelihood estimation problem. Using an expectation-maximization approach, we demonstrate that a reconstruction algorithm can be obtained following a simple substitution rule from the one previously derived without electronic noise considerations. To illustrate the applicability of the substitution rule, we present examples of a fully iterative reconstruction algorithm and a sinogram smoothing algorithm both in transmission CT reconstruction when the measured signal contains additive electronic noise. Our simulation studies show the potential usefulness of accurate electronic noise modeling in low-dose CT applications.

A faster ordered-subset convex algorithm for iterative reconstruction in a rotation-free micro-CT system

We present a faster iterative reconstruction algorithm based on the ordered-subset convex (OSC) algorithm for transmission CT. The OSC algorithm was modified such that it calculates the normalization term before the iterative process in order to save computational cost. The modified version requires only one backprojection per iteration as compared to two required for the original OSC. We applied the modified OSC (MOSC) algorithm to a rotation-free micro-CT system that we proposed previously, observed its performance, and compared with the OSC algorithm for 3D cone-beam reconstruction. Measurements on the reconstructed images as well as the point spread functions show that MOSC is quite similar to OSC; in noise-resolution trade-off, MOSC is comparable with OSC in a regular-noise situation and it is slightly worse than OSC in an extremely high-noise situation. The timing record shows that MOSC saves 25-30% CPU time, depending on the number of iterations used. We conclude that the MOSC algorithm is more efficient than OSC and provides comparable images.

Development of novel statistical reconstruction algorithms for poly-energetic X-ray computed tomography

A beam-hardening effect is a common problem affecting the quantitative aspects of X-ray computed tomography (CT). We have developed two statistical reconstruction algorithms for poly-energetic X-ray CT that can effectively reduce the beam-hardening effect. Phantom tests were used to evaluate our approach in comparison with traditional correction methods. Unlike previous methods, our algorithm utilizes multiple energy-corresponding blank scans to estimate the attenuation map for a particular energy spectrum. Therefore, our algorithm is an energy-selective reconstruction. In addition to benefits over other statistical algorithms for poly-energetic reconstruction, our algorithm has the advantage of not requiring prior knowledge of the object material, the energy spectrum of the source and the energy sensitivity of the detector. The results showed an improvement in coefficient of variation, uniformity and signal-to-noise ratio; overall, this novel approach produces a better beam-hardening correction.

Motion-compensated iterative cone-beam CT image reconstruction with adapted blobs as basis functions

This paper presents a three-dimensional method to reconstruct moving objects from cone-beam X-ray projections using an iterative reconstruction algorithm and a given motion vector field. For the image representation, adapted blobs are used, which can be implemented efficiently as basis functions. Iterative reconstruction requires the calculation of line integrals (forward projections) through the image volume, which are compared with the actual measurements to update the image volume. In the existence of a divergent motion vector field, a change in the volumes of the blobs has to be taken into account in the forward and backprojections. An efficient method to calculate the line integral through the adapted blobs is proposed. It solves the problem, how to compensate for the divergence in the motion vector field on a grid of basis functions. The method is evaluated on two phantoms, which are subject to three different known motions. Moreover, a motion-compensated filtered back-projection reconstruction method is used, and the reconstructed images are compared. Using the correct motion vector field with the iterative motion-compensated reconstruction, sharp images are obtained, with a quality that is significantly better than gated reconstructions.

High temporal resolution and streak-free four-dimensional cone-beam computed tomography

Cone-beam computed tomography (CBCT) has been clinically used to verify patient position and to localize the target of treatment in image-guided radiation therapy (IGRT). However, when the chest and the upper abdomen are scanned, respiratory-induced motion blurring limits the utility of CBCT. In order to mitigate this blurring, respiratory-gated CBCT, i.e. 4D CBCT, was introduced. In 4D CBCT, the cone-beam projection data sets acquired during a gantry rotation are sorted into several respiratory phases. In these gated reconstructions, the number of projections for each respiratory phase is significantly reduced. Consequently, undersampling streaking artifacts are present in the reconstructed images, and the image contrast resolution is also significantly compromised. In this paper, we present a new method to simultaneously achieve both high temporal resolution ( approximately 100 ms) and streaking artifact-free image volumes in 4D CBCT. The enabling technique is a newly proposed image reconstruction method, i.e. prior image constrained compressed sensing (PICCS), which enables accurate image reconstruction using vastly undersampled cone-beam projections and a fully sampled prior image. Using PICCS, a streak-free image can be reconstructed from 10-20 cone-beam projections while the signal-to-noise ratio is determined by a denoising feature of the selected objective function and by the prior image, which is reconstructed using all of the acquired cone-beam projections. This feature of PICCS breaks the connection between the temporal resolution and streaking artifacts' level in 4D CBCT. Numerical simulations and experimental phantom studies have been conducted to validate the method.

Binary encoding of multiplexed images in mixed noise

Binary coding of multiplexed signals and images has been studied in the context of spectroscopy with models of either purely constant or purely proportional noise, and has been shown to result in improved noise performance under certain conditions. We consider the case of mixed noise in an imaging system consisting of multiple individually-controllable sources (X-ray or near-infrared, for example) shining on a single detector. We develop a mathematical model for the noise in such a system and show that the noise is dependent on the properties of the binary coding matrix and on the average number of sources used for each code. Each binary matrix has a characteristic linear relationship between the ratio of proportional-to-constant noise and the noise level in the decoded image. We introduce a criterion for noise level, which is minimized via a genetic algorithm search. The search procedure results in the discovery of matrices that outperform the Hadamard S-matrices at certain levels of mixed noise. Simulation of a seven-source radiography system demonstrates that the noise model predicts trends and rank order of performance in regions of nonuniform images and in a simple tomosynthesis reconstruction. We conclude that the model developed provides a simple framework for analysis, discovery, and optimization of binary coding patterns used in multiplexed imaging systems.

Digital tomosynthesis aided by low-resolution exact computed tomography

Tomosynthesis reconstructs 3-dimensional images of an object from a significantly fewer number of projections as compared with that required by computed tomography (CT). A major problem with tomosynthesis is image artifacts associated with the data incompleteness. In this article, we propose a hybrid tomosynthesis approach to achieve higher image quality as compared with competing methods. In this approach, a low-resolution CT scan is followed by a high-resolution tomosynthesis scan. Then, both scans are combined to reconstruct images. To evaluate the image quality of the proposed method, we design a new breast phantom for numerical simulation and physical experiments. The results show that images obtained by our approach are clearly better than those obtained without such a CT scan.

A three-dimensional statistical approach to improved image quality for multislice helical CT

Multislice helical computed tomography scanning offers the advantages of faster acquisition and wide organ coverage for routine clinical diagnostic purposes. However, image reconstruction is faced with the challenges of three-dimensional cone-beam geometry, data completeness issues, and low dosage. Of all available reconstruction methods, statistical iterative reconstruction (IR) techniques appear particularly promising since they provide the flexibility of accurate physical noise modeling and geometric system description. In this paper, we present the application of Bayesian iterative algorithms to real 3D multislice helical data to demonstrate significant image quality improvement over conventional techniques. We also introduce a novel prior distribution designed to provide flexibility in its parameters to fine-tune image quality. Specifically, enhanced image resolution and lower noise have been achieved, concurrently with the reduction of helical cone-beam artifacts, as demonstrated by phantom studies. Clinical results also illustrate the capabilities of the algorithm on real patient data. Although computational load remains a significant challenge for practical development, superior image quality combined with advancements in computing technology make IR techniques a legitimate candidate for future clinical applications.

Majorization-minimization algorithms for wavelet-based image restoration

Standard formulations of image/signal deconvolution under wavelet-based priors/regularizers lead to very high-dimensional optimization problems involving the following difficulties: the non-Gaussian (heavy-tailed) wavelet priors lead to objective functions which are nonquadratic, usually nondifferentiable, and sometimes even nonconvex; the presence of the convolution operator destroys the separability which underlies the simplicity of wavelet-based denoising. This paper presents a unified view of several recently proposed algorithms for handling this class of optimization problems, placing them in a common majorization-minimization (MM) framework. One of the classes of algorithms considered (when using quadratic bounds on nondifferentiable log-priors) shares the infamous "singularity issue" (SI) of "iteratively reweighted least squares" (IRLS) algorithms: the possibility of having to handle infinite weights, which may cause both numerical and convergence issues. In this paper, we prove several new results which strongly support the claim that the SI does not compromise the usefulness of this class of algorithms. Exploiting the unified MM perspective, we introduce a new algorithm, resulting from using l1 bounds for nonconvex regularizers; the experiments confirm the superior performance of this method, when compared to the one based on quadratic majorization. Finally, an experimental comparison of the several algorithms, reveals their relative merits for different standard types of scenarios.

Application of boundary detection information in breast tomosynthesis reconstruction

Digital tomosynthesis mammography (DTM) is one of the most promising techniques that can potentially improve early detection of breast cancers. DTM can provide three-dimensional (3D) structural information by reconstructing the whole imaged volume from a sequence of projection-view (PV) mammograms that are acquired at a small number of projection angles over a limited angular range. Our previous study showed that simultaneous algebraic reconstruction technique (SART) can produce satisfactory tomosynthesized image quality compared to maximum likelihood-type algorithms. To improve the efficiency of DTM reconstruction and address the problem of boundary artifacts, we have developed methods to incorporate both two-dimensional (2D) and 3D breast boundary information within the SART reconstruction algorithm in this study. A second generation GE prototype tomosynthesis mammography system with a stationary digital detector was used for PV image acquisition from 21 angles in 3 degrees increments over a +/- 30 degrees angular range. The 2D breast boundary curves on all PV images were obtained by automated segmentation and were used to restrict the SART reconstruction to be performed only within the breast volume. The computation time of SART reconstruction was reduced by 76.3% and 69.9% for cranio-caudal and mediolateral oblique views, respectively, for the chosen example. In addition, a 3D conical trimming method was developed in which the 2D breast boundary curves from all PVs were back projected to generate the 3D breast surface. This 3D surface was then used to eliminate the multiple breast shadows outside the breast volume due to reconstruction by setting these voxels to a constant background value. Our study demonstrates that, by using the 2D and 3D breast boundary information, all breast boundary and most detector boundary artifacts can be effectively removed on all tomosynthesized slices.

Alternating minimization algorithms for transmission tomography

A family of alternating minimization algorithms for finding maximum-likelihood estimates of attenuation functions in transmission X-ray tomography is described. The model from which the algorithms are derived includes polyenergetic photon spectra, background events, and nonideal point spread functions. The maximum-likelihood image reconstruction problem is reformulated as a double minimization of the I-divergence. A novel application of the convex decomposition lemma results in an alternating minimization algorithm that monotonically decreases the objective function. Each step of the minimization is in closed form. The family of algorithms includes variations that use ordered subset techniques for increasing the speed of convergence. Simulations demonstrate the ability to correct the cupping artifact due to beam hardening and the ability to reduce streaking artifacts that arise from beam hardening and background events.

Estrogen therapy and brain muscarinic receptor density in healthy females: a SPET study

Estrogen Therapy (ET) may protect against age-related cognitive decline and neuropsychiatric disorders (e.g. Alzheimer's disease). The biological basis for this putative neuroprotective effect is not fully understood, but may include modulation of cholinergic systems. Cholinergic dysfunction has been implicated in age-related memory impairment and Alzheimer's disease. However, to date no one has investigated the effect of long-term ET on brain cholinergic muscarinic receptor aging, and related this to cognitive function. We used Single Photon Emission Tomography (SPET) and (R,R)[(123)I]-I-QNB, a novel ligand with high affinity for m(1)/m(4) muscarinic receptors, to examine the effect of long-term ET and age on brain m(1)/m(4) receptors in healthy females. We included 10 younger premenopausal subjects and 22 postmenopausal women; 11 long-term ET users (all treated following surgical menopause) and 11 ET never-users (surgical menopause, n=2). Also, verbal memory and executive function was assessed in all postmenopausal subjects. Compared to young women, postmenopausal women (ET users and never-users combined) had significantly lower muscarinic receptor density in all brain regions examined. ET users also had higher muscarinic receptor density than ET never-users in all the brain regions, and this reached statistical significance in left striatum and hippocampus, lateral frontal cortex and thalamus. Moreover, in ET users, (R,R)[(123)I]-I-QNB binding in left hippocampus and temporal cortex was significantly positively correlated with plasma estradiol levels. We also found evidence for improved executive function in ET users as compared to ET never-users. However, there was no significant relationship between receptor binding and cognitive function within any of the groups. In healthy postmenopausal women use of long-term ET is associated with reduced age-related differences in muscarinic receptor binding, and this may be related to serum estradiol levels.

Noise and resolution in images reconstructed with FBP and OSC algorithms for CT

This paper presents a comparison between an analytical and a statistical iterative reconstruction algorithm for computed transmission tomography concerning their noise and resolution performance. The reconstruction of two-dimensional images from simulated fan-beam transmission data is performed with a filtered back-projection (FBP) type reconstruction and an iterative ordered subsets convex (OSC) maximum-likelihood method. A special software phantom, which allows measuring the resolution and noise in a nonambiguous way, is used to simulate transmission tomography scans with different signal-to-noise ratios (SNR). The noise and modulation transfer function is calculated for FBP and OSC reconstruction at several positions, distributed over the field-of-view (FOV). The reconstruction with OSC using different numbers of subsets shows an inverse linear relation to the number of iterations that are necessary to reach a certain resolution and SNR, i.e., increasing the number of subsets by a factor x reduces the number of required iterations by the same factor. The OSC algorithm is able to achieve a nearly homogeneous high resolution over the whole FOV, which is not achieved with FBP. The OSC method achieves a lower level of noise compared with FBP at the same resolution. The reconstruction with OSC can save a factor of up to nine of x-ray dose compared with FBP in the investigated range of noise levels.

A comparative study of limited-angle cone-beam reconstruction methods for breast tomosynthesis

Digital tomosynthesis mammography (DTM) is a promising new modality for breast cancer detection. In DTM, projection-view images are acquired at a limited number of angles over a limited angular range and the imaged volume is reconstructed from the two-dimensional projections, thus providing three-dimensional structural information of the breast tissue. In this work, we investigated three representative reconstruction methods for this limited-angle cone-beam tomographic problem, including the backprojection (BP) method, the simultaneous algebraic reconstruction technique (SART) and the maximum likelihood method with the convex algorithm (ML-convex). The SART and ML-convex methods were both initialized with BP results to achieve efficient reconstruction. A second generation GE prototype tomosynthesis mammography system with a stationary digital detector was used for image acquisition. Projection-view images were acquired from 21 angles in 3 degrees increments over a +/- 30 degrees angular range. We used an American College of Radiology phantom and designed three additional phantoms to evaluate the image quality and reconstruction artifacts. In addition to visual comparison of the reconstructed images of different phantom sets, we employed the contrast-to-noise ratio (CNR), a line profile of features, an artifact spread function (ASF), a relative noise power spectrum (NPS), and a line object spread function (LOSF) to quantitatively evaluate the reconstruction results. It was found that for the phantoms with homogeneous background, the BP method resulted in less noisy tomosynthesized images and higher CNR values for masses than the SART and ML-convex methods. However, the two iterative methods provided greater contrast enhancement for both masses and calcification, sharper LOSF, and reduced interplane blurring and artifacts with better ASF behaviors for masses. For a contrast-detail phantom with heterogeneous tissue-mimicking background, the BP method had strong blurring artifacts along the x-ray source motion direction that obscured the contrast-detail objects, while the other two methods can remove the superimposed breast structures and significantly improve object conspicuity. With a properly selected relaxation parameter, the SART method with one iteration can provide tomosynthesized images comparable to those obtained from the ML-convex method with seven iterations, when BP results were used as initialization for both methods.

Pose estimation of known objects during transmission tomographic image reconstruction

We address the problem of image formation in transmission tomography when metal objects of known composition and shape, but unknown pose, are present in the scan subject. Using an alternating minimization (AM) algorithm, derived from a model in which the detected data are viewed as Poisson-distributed photon counts, we seek to eliminate the streaking artifacts commonly seen in filtered back projection images containing high-contrast objects. We show that this algorithm, which minimizes the I-divergence (or equivalently, maximizes the log-likelihood) between the measured data and model-based estimates of the means of the data, converges much faster when knowledge of the high-density materials (such as brachytherapy applicators or prosthetic implants) is exploited. The algorithm incorporates a steepest descent-based method to find the position and orientation (collectively called the pose) of the known objects. This pose is then used to constrain the image pixels to their known attenuation values, or, for example, to form a mask on the "missing" projection data in the shadow of the objects. Results from two-dimensional simulations are shown in this paper. The extension of the model and methods used to three dimensions is outlined.

[123I]TPCNE--a novel SPET tracer for the sigma-1 receptor: first human studies and in vivo haloperidol challenge

[123I]TPCNE (1(trans-[123I]iodopropen-2-yl)-4-[(4-cyanophenoxy)methyl] piperidine; Ki = 0.67 nM; log P = 3.36) is a novel sigma-1 receptor SPET ligand. In this study, we developed an optimized labeling method for [123I]TPCNE and investigated the kinetics, binding characteristics, and whole-body distribution of this tracer for the first time in humans. We also performed a challenge with the sigma-1 receptor antagonist haloperidol against [123I]TPCNE. Seven healthy volunteers were recruited. Dynamic brain SPET scans were performed following i.v. administration of 185 MBq [123I]TPCNE in all seven subjects. Three of the subjects were given oral haloperidol (2.5 mg) approximately 1 h before the scan. The dynamic data were analyzed with both reversible and irreversible compartmental models.[123I]TPCNE showed high uptake in brain and liver. All non-haloperidol-treated subjects showed a high whole-brain uptake (average: 8.7% of injected activity). No significant clearance of the tracer was seen up to 30 h post injection. In the haloperidol-treated subjects, the time-activity curves clearly demonstrated clearance of the tracer from the brain. Regional radioactivity concentrations were reduced by haloperidol from 42% in the cerebellum to 73% in the thalamus.[(123)I]TPCNE demonstrated high brain uptake, with highest binding found in the posterior cingulate. A region in which binding was unaffected by haloperidol pretreatment could not be identified, and the time-activity data were best described by an irreversible model.

Rectification for cone-beam projection and backprojection

The purpose of this paper is to derive a technique for accelerating the computation of cone-beam forward and backward projections that are the basic steps of tomographic reconstruction. The cone-beam geometry of C-arm systems is commonly described with projection matrices. Such matrices provide a continuous framework for analyzing the flow of operations needed to compute backprojection for analytical reconstruction, as well as the combination of forward and backward projections for iterative reconstruction. The proposed rectification technique resampies the original data to planes that are aligned with two of the reconstructed volume main axes, so that the original cone-beam geometry can be replaced by a simpler geometry, where succession of plane magnifications are involved only. Rectification generalizes previous independent results to the cone-beam backprojection of preprocessed data as well as to cone-beam iterative reconstruction. The memory access pattern of simple magnifications provides superior predictability and is, therefore, easier to optimize, independently of the choice of the interpolation technique. Rectification is also shown to provide control over interpolation errors through oversampling, allowing tradeoffs between computation speed and precision to be set. Experimental results are provided for linear and nearest neighbor interpolations, based on simulations, as well as phantom and patient data acquired on a digital C-arm system.

Convergent incremental optimization transfer algorithms: application to tomography

No convergent ordered subsets (OS) type image reconstruction algorithms for transmission tomography have been proposed to date. In contrast, in emission tomography, there are two known families of convergent OS algorithms: methods that use relaxation parameters, and methods based on the incremental expectation-maximization (EM) approach. This paper generalizes the incremental EM approach by introducing a general framework, "incremental optimization transfer." The proposed algorithms accelerate convergence speeds and ensure global convergence without requiring relaxation parameters. The general optimization transfer framework allows the use of a very broad family of surrogate functions, enabling the development of new algorithms. This paper provides the first convergent OS-type algorithm for (nonconcave) penalized-likelihood (PL) transmission image reconstruction by using separable paraboloidal surrogates (SPS) which yield closed-form maximization steps. We found it is very effective to achieve fast convergence rates by starting with an OS algorithm with a large number of subsets and switching to the new "transmission incremental optimization transfer (TRIOT)" algorithm. Results show that TRIOT is faster in increasing the PL objective than nonincremental ordinary SPS and even OS-SPS yet is convergent.

A new simple iterative reconstruction algorithm for SPECT transmission measurement

This paper proposes a new iterative reconstruction algorithm for transmission tomography and compares this algorithm with several other methods. The new algorithm is simple and resembles the emission ML-EM algorithm in form. Due to its simplicity, it is easy to implement and fast to compute a new update at each iteration. The algorithm also always guarantees non-negative solutions. Evaluations are performed using simulation studies and real phantom data. Comparisons with other algorithms such as convex, gradient, and logMLEM show that the proposed algorithm is as good as others and performs better in some cases.

Measuring SSRI occupancy of SERT using the novel tracer [123I]ADAM: a SPECT validation study

PURPOSE

Serotonergic brain regions play a crucial role in the modulation of emotion, and serotonergic dysfunction may contribute to several neurological disorders. [123I]ADAM is a novel SPECT tracer which binds with high affinity to serotonin transporters (SERT). The objective of this study was to compare different methods for the quantification of tracer binding and to develop a simplified single-scan protocol for this tracer, as well as to investigate its potential for characterisation of the transporter occupancy versus plasma concentration curve of a selective serotonin re-uptake inhibitor (SSRI).

METHODS

Dynamic SPECT scans were performed on 16 healthy volunteers after administration of approximately 150 MBq [123I]ADAM. Data were acquired from the time of injection until approximately 5.5 h after injection in 30- or 45-min sessions. Each subject was scanned twice: with and without pre-treatment with the SSRI citalopram in various dosage regimens. The plasma concentration of citalopram (C(p)) was determined from venous samples. Images were reconstructed by filtered back-projection with scatter and attenuation correction. Tracer binding was quantified for midbrain, striatum and thalamus using cerebellum as a reference region. Quantification was done by kinetic modelling, graphical analysis and multi-linear regression, as well as by the ratio method, with binding potential (BP2) as the outcome measure. The SERT occupancy by citalopram was determined relative to the baseline scan for each subject, and the occupancy versus C(p) curve was fitted with the E(max) model.

RESULTS

The highest binding of [123I]ADAM was in midbrain (mean baseline BP2+/-SD=1.31+/-0.29), with lower binding in thalamus (0.79+/-0.16) and striatum (0.66+/-0.13). There was good agreement between BP2 values obtained by different quantification methods. Using the ratio method, the best agreement with kinetic modelling was obtained with data from the time interval [200,260] min after injection. The fitting of the midbrain occupancy curve yielded a maximum occupancy of 84% and a plasma concentration required to reach 50% of the maximum of 2.5 ng/ml, with a goodness-of-fit variability of 13% (SD).

CONCLUSION

Binding of [123I]ADAM to SERT in midbrain can be quantified with a single scan starting 200 min after injection. However, the variability of estimated occupancy values may be too high for critical assessment of occupancy of SERT by SSRI.

Accelerated penalized weighted least-squares and maximum likelihood algorithms for reconstructing transmission images from PET transmission data

We present penalized weighted least-squares (PWLS) and penalized maximum-likelihood (PML) methods for reconstructing transmission images from positron emission tomography transmission data. First, we view the problem of minimizing the weighted least-squares (WLS) and maximum likelihood objective functions as a sequence of nonnegative least-squares minimization problems. This viewpoint follows from using certain quadratic functions as surrogate functions for the WLS and maximum likelihood objective functions. Second, we construct surrogate functions for a class of penalty functions that yield closed form expressions for the iterates of the PWLS and PML algorithms. Due to the slow convergence of the PWLS and PML algorithms, accelerated versions of them are developed that are theoretically guaranteed to monotonically decrease their respective objective functions. In experiments using real phantom data, the PML images produced the most accurate attenuation correction factors. On the other hand, the PWLS images produced images with the highest levels of contrast for low-count data.

SPET imaging of central muscarinic receptors with (R,R)[123I]-I-QNB: methodological considerations

Investigations on the effect of normal healthy ageing on the muscarinic system have shown conflicting results. Also, in vivo determination of muscarinic receptor binding has been hampered by a lack of subtype selective ligands and differences in methods used for quantification of receptor densities. Recent in vitro and in vivo work with the muscarinic antagonist (R,R)-I-QNB indicates this ligand has selectivity for m(1) and m(4) muscarinic receptor subtypes. Therefore, we used (R,R)[(123)I]-I-QNB and single photon emission tomography to study brain m(1) and m(4) muscarinic receptors in 25 healthy female subjects (11 younger subjects, age range 26-32 years and 14 older subjects, age range 57-82 years). Our aims were to ascertain the viability of tracer administration and imaging within the same day, and to evaluate whether normalization to whole brain, compared to normalization to cerebellum, could alter the clinical interpretation of results. Images were analyzed using the simplified reference tissue model and by two ratio methods: normalization to whole brain and normalization to cerebellum. Significant correlations were observed between kinetic analysis and normalization to cerebellum, but not to whole brain. Both the kinetic analysis and normalization to cerebellum showed age-related reductions in muscarinic binding in frontal, orbitofrontal, and parietal regions. Normalization to whole brain, however, failed to detect age-related changes in any region. Here we show that, for this radiotracer, normalizing to a region of negligible specific binding (cerebellum) significantly improves sensitivity when compared to global normalization.

A bolus/infusion paradigm for the novel NMDA receptor SPET tracer [123i]CNS 1261

We have previously performed quantitative kinetic modeling of [(123)I]CNS 1261, a new SPET ligand for the MK801 intrachannel site of the NMDA receptor. We now report a bolus-infusion protocol, which eliminates the need for arterial blood sampling. Dynamic SPET scanning and venous blood sampling were performed in 7 healthy volunteers. Good agreement was obtained between kinetic and equilibrium analysis. SPET scanning with a bolus-infusion protocol is a valid method to estimate the total volume of distribution for [(123)I]CNS 1261 in clinical populations.

Analysis of D2 dopamine receptor occupancy with quantitative SPET using the high-affinity ligand [123I]epidepride: resolving conflicting findings

Recent studies of limbic cortical dopamine D(2) receptor occupancy by clozapine using high-affinity PET and SPET radioligands have produced conflicting findings. It has been suggested that these divergent findings are due to between-study differences in the method used to estimate D(2) receptor-binding potential. We compared different methods for estimating striatal and temporal cortical D(2) receptor occupancy with high-affinity tracers. In vivo experimental SPET data, obtained with [(123)I]epidepride were analysed with reference tissue kinetic modeling and with the ratio method, applied to data corresponding to short (60 min) and long (240 min) acquisition times. Dopamine D(2) receptor occupancy by the atypical antipsychotic drug risperidone was evaluated. Simulation experiments were also performed, comparing occupancy values obtained for different receptor densities in relation to different data acquisition times. The simulation results revealed that previously published data regarding errors in occupancy estimation by analysis of time activity data acquired for 60 min cannot be extrapolated to studies performed over 240 min. The ratio method provided accurate temporal cortical D(2) receptor occupancy values when applied to data from a late time period, but underestimated the occupancy with earlier data. In striatum, both the late data ratio method and reference tissue kinetic modeling using all data underestimated D(2) receptor occupancy. However, more accurate analyses of striatal D(2) occupancy still showed selective limbic/cortical occupancy by risperidone. Our results substantiate the previous [(123)I]epidepride findings of high temporal cortical occupancy by other atypical antipsychotic drugs and suggest that a potential source of conflicting findings might be short scanning times imposed by [(11)C]FLB 457, leading to underestimation of temporal cortical D(2) receptor occupancy by this method.

Kinetic modelling of [123I]CNS 1261--a potential SPET tracer for the NMDA receptor

N-(1-napthyl)-N'-(3-[(123)I]-iodophenyl)-N-methylguanidine ([(123)I]CNS 1261) is a novel SPET ligand developed for imaging the NMDA receptor intra-channel MK 801/PCP/ketamine site. Data was acquired in 7 healthy volunteers after bolus injection of [(123)I]CNS 1261. Kinetic modeling showed reversible tracer binding. Arterial and venous time-activity curves overlapped after 90 min. The rank order of binding was: Thalamus > striatum > cortical regions > white matter. This distribution concurs with [(11)C]-ketamine and [(18)F]-memantine PET studies. These data provide a methodological basis for further direct in vivo challenge studies.

Tomographic mammography using a limited number of low-dose cone-beam projection images

A method is described for using a limited number (typically 10-50) of low-dose radiographs to reconstruct the three-dimensional (3D) distribution of x-ray attenuation in the breast. The method uses x-ray cone-beam imaging, an electronic digital detector, and constrained nonlinear iterative computational techniques. Images are reconstructed with high resolution in two dimensions and lower resolution in the third dimension. The 3D distribution of attenuation that is projected into one image in conventional mammography can be separated into many layers (typically 30-80 1-mm-thick layers, depending on breast thickness), increasing the conspicuity of features that are often obscured by overlapping structure in a single-projection view. Schemes that record breast images at nonuniform angular increments, nonuniform image exposure, and nonuniform detector resolution are investigated in order to reduce the total x-ray exposure necessary to obtain diagnostically useful 3D reconstructions, and to improve the quality of the reconstructed images for a given exposure. The total patient radiation dose can be comparable to that used for a standard two-view mammogram. The method is illustrated with images from mastectomy specimens, a phantom, and human volunteers. The results show how image quality is affected by various data-collection protocols.

Optimizing limbic selective D2/D3 receptor occupancy by risperidone: a [123I]-epidepride SPET study

Selective action at limbic cortical dopamine D2-like receptors is a putative mechanism of atypical antipsychotic efficacy with few extrapyramidal side effects. Although risperidone is an atypical antipsychotic with high affinity for D2 receptors, low-dose risperidone treatment is effective without inducing extrapyramidal symptoms. The objective was to test the hypothesis that treatment with low-dose risperidone results in 'limbic selective' D2/D3 receptor blockade in vivo. Dynamic single photon emission tomography (SPET) sequences were obtained over 5 hours after injection of [123I]-epidepride (approximately 150 MBq), using a high-resolution triple-headed brain scanner (Marconi Prism 3000XP). Kinetic modelling was performed using the simplified reference region model to obtain binding potential values. Estimates of receptor occupancy were made relative to a normal volunteer control group (n = 5). Six patients treated with low-dose risperidone (mean = 2.6 mg) showed moderate levels of D2/D3 occupancy in striatum (49.9%), but higher levels of D2/D3 occupancy in thalamus (70.8%) and temporal cortex (75.2%). Occupancy values in striatum were significantly different from thalamus (F (1,4) = 26.3, p < 0.01) and from temporal cortex (F (1,4) = 53.4, p < 0.01). This is the first study to evaluate striatal and extrastriatal occupancy of risperidone. Low dose treatment with risperidone achieves a similar selectivity of limbic cortical over striatal D2/D3 receptor blockade to that of atypical antipsychotics with lower D2/D3 affinity such as clozapine, olanzapine and quetiapine. This finding is consistent with the relevance of 'limbic selective' D2/D3 receptor occupancy to the therapeutic efficacy of atypical antipsychotic drugs.