Population analysis of brain morphology from magnetic resonance images contributes to the study and understanding of neurological diseases. Such analysis typically involves segmentation of a large set of images and comparisons of these segmentations between relevant subgroups of images (e.g., "normal" versus "diseased"). The images of each subgroup are usually selected in advance in a supervised way based on clinical knowledge. Their segmentations are typically guided by one or more available atlases, assumed to be suitable for the images at hand. We present a data-driven probabilistic framework that simultaneously performs atlas-guided segmentation of a heterogeneous set of brain MR images and clusters the images in homogeneous subgroups, while constructing separate probabilistic atlases for each cluster to guide the segmentation. The main benefits of integrating segmentation, clustering and atlas construction in a single framework are that: 1) our method can handle images of a heterogeneous group of subjects and automatically identifies homogeneous subgroups in an unsupervised way with minimal prior knowledge, 2) the subgroups are formed by automatical detection of the relevant morphological features based on the segmentation, 3) the atlases used by our method are constructed from the images themselves and optimally adapted for guiding the segmentation of each subgroup, and 4) the probabilistic atlases represent the morphological pattern that is specific for each subgroup and expose the groupwise differences between different subgroups. We demonstrate the feasibility of the proposed framework and evaluate its performance with respect to image segmentation, clustering and atlas construction on simulated and real data sets including the publicly available BrainWeb and ADNI data. It is shown that combined segmentation and atlas construction leads to improved segmentation accuracy. Furthermore, it is demonstrated that the clusters generated by our unsupervised framework largely coincide with the clinically determined subgroups in case of disease-specific differences in brain morphology and that the differences between the cluster-specific atlases are in agreement with the expected disease-specific patterns, indicating that our method is capable of detecting the different modes in a population. Our method can thus be seen as a comprehensive image-driven population analysis framework that can contribute to the detection of novel subgroups and distinctive image features, potentially leading to new insights in the brain development and disease.

Research in face recognition has continuously been challenged by extrinsic (head pose, lighting conditions) and intrinsic (facial expression, aging) sources of variability. While many survey papers on face recognition exist, in this paper, we focus on a comparative study of 3-D face recognition under expression variations. As a first contribution, 3-D face databases with expressions are listed, and the most important ones are briefly presented and their complexity is quantified using the iterative closest point (ICP) baseline recognition algorithm. This allows to rank the databases according to their inherent difficulty for face-recognition tasks. This analysis reveals that the FRGC v2 database can be considered as the most challenging because of its size, the presence of expressions and outliers, and the time lapse between the recordings. Therefore, we recommend to use this database as a reference database to evaluate (expression-invariant) 3-D face-recognition algorithms. We also determine and quantify the most important factors that influence the performance. It appears that performance decreases 1) with the degree of nonfrontal pose, 2) for certain expression types, 3) with the magnitude of the expressions, 4) with an increasing number of expressions, and 5) for a higher number of gallery subjects. Future 3-D face-recognition algorithms should be evaluated on the basis of all these factors. As the second contribution, a survey of published 3-D face-recognition methods that deal with expression variations is given. These methods are subdivided into three classes depending on the way the expressions are handled. Region-based methods use expression-stable regions only, while other methods model the expressions either using an isometric or a statistical model. Isometric models assume the deformation because of expression variation to be (locally) isometric, meaning that the deformation preserves lengths along the surface. Statistical models learn how the facial soft tissue deforms during expressions based on a training database with expression labels. Algorithmic performances are evaluated by the comparison of recognition rates for identification and verification. No statistical significant differences in class performance are found between any pair of classes.

In this paper the problem of pairwise model-to-scene point set registration is considered. Three contributions are made. Firstly, the relations between correspondence-based and some information-theoretic point cloud registration algorithms are formalized. Starting from the observation that the outlier handling of existing methods relies on heuristically determined models, a second contribution is made exploiting aforementioned relations to derive a new robust point set registration algorithm. Representing model and scene point clouds by mixtures of Gaus-sians, the method minimizes their Kullback-Leibler divergence both w.r.t. the registration transformation parameters and w.r.t. the scene's mixture coefficients. This results in an Expectation-Maximization Iterative Closest Point (EM-ICP) approach with a parameter-free outlier model that is optimal in information-theoretical sense. While the current (CUDA) implementation is limited to the rigid registration case, the underlying theory applies to both rigid and non-rigid point set registration. As a by-product of the registration algorithm's theory, a third contribution is made by suggesting a new point cloud Kernel Density Estimation approach which relies on maximizing the resulting distribution's entropy w.r.t. the kernel weights. The rigid registration algorithm is applied to align different patches of the publicly available Stanford Dragon and Stanford Happy Budha range data. The results show good performance regarding accuracy, robustness and convergence range.

Since most 3D cameras cannot capture the complete 3D face, an important challenge in 3D face recognition is the comparison of two 3D facial surfaces with little or no over lap. In this paper, a local feature method is presented to tackle this challenge exploiting the symmetry of the human face. Features are located and described using an extension of SIFT for meshes (meshSIFT). As such, features are localized as extrema in the curvature scale space of the in put mesh, and are described by concatenating histograms of shape indices and slant angles of the neighborhood. For 3D face scans with sufficient overlap, the number of matching meshSIFT features is a reliable measure for face recognition purposes. However, as the feature descriptor is not symmetrical, features on one face are not matched with their symmetrical counterpart on another face impeding their feasibility for comparison efface scans with limited or no (left-right) overlap. In order to alleviate this problem, facial symmetry could be used to increase the overlap between two face scans by mirroring one of both faces w.r.t. an arbitrary plane. As this would increase the computational demand, this paper proposes an efficient approach to de scribe the features of a mirrored face by mirroring the mesh SIFT descriptors of the input face. The presented method is validated on the data of the "SHREC '11: Face Scans" contest, containing many partial scans. This resulted in a recognition rate of 98.6% and a mean average precision of 93.3%, clearly outperforming all other participants in the challenge.

In this paper a method is presented for the automated identification of cephalometric anatomical landmarks in craniofacial cone-beam CT images. This method makes use of statistical models, incorporating both local appearance and shape knowledge obtained from training data. Firstly, the local appearance model captures the local intensity pattern around each anatomical landmark in the image. Secondly, the shape model contains a local and a global component. The former improves the flexibility, whereas the latter improves the robustness of the algorithm. Using a leave-one-out approach to the training data, we assess the overall accuracy of the method. The mean and median error values for all landmarks are equal to 2.55 mm and 1.72 mm, respectively.

The recognition of faces under varying expressions is one of the current challenges in the face recognition community. In this paper, we propose a method fusing different complementary approaches each dealing with expression variations. The first approach uses an isometric deformation model and is based on the largest singular values of the geodesic distance matrix as an expression-invariant shape descriptor. The second approach performs recognition on the more rigid parts of the face that are less affected by expression variations. Several fusion techniques are examined for combining the approaches. The presented method is validated on a subset of 900 faces of the BU-3DFE face database resulting in an equal error rate of 5.85% for the verification scenario and a rank 1 recognition rate of 94.48% for the identification scenario using the sum rule as fusion technique. This result outperforms other 3D expression-invariant face recognition methods on the same database.

Magnetic resonance (MR) cine images are often used to clinically assess left ventricular cardiac function. In a typical study, multiple 2-D long axis (LA) and short axis (SA) cine images are acquired, each in a different breath-hold. Differences in lung volume during breath-hold and overall patient motion distort spatial alignment of the images thus complicating spatial integration of all image data in three dimensions. We present a fully automatic postprocessing approach to correct these slice misalignments. The approach is based on the constrained optimization of the intensity similarity of intersecting image lines after the automatic definition of a region of interest. It uses all views and all time frames simultaneously. Our method models both in-plane and out-of-plane translations and full 3-D rotations, can be applied retrospectively and does not require a cardiac wall segmentation. The method was validated on both healthy volunteer and patient data with simulated misalignments, as well as on clinical multibreath-hold patient data. For the simulated data, subpixel accuracy could be obtained using translational correction. The possibilities and limitations of rotational correction were investigated and discussed. For the clinical multibreath-hold patient data sets, the median discrepancy between manual SA and LA contours was reduced from 2.83 to 1.33 mm using the proposed correction method. We have also shown the usefulness of the correction method for functional analysis on clinical image data. The same clinical multibreath-hold data sets were resegmented after positional correction, taking newly available complementary information of intersecting slices into account, further reducing the median discrepancy to 0.43 mm. This is due to the integration of the 2-D slice information into 3-D space.

Maximization of mutual information (MMI) is a popular similarity measure for medical image registration. Although its accuracy and robustness has been demonstrated for rigid body image registration, extending MMI to nonrigid image registration is not trivial and an active field of research. We propose conditional mutual information (cMI) as a new similarity measure for nonrigid image registration. cMI starts from a 3-D joint histogram incorporating, besides the intensity dimensions, also a spatial dimension expressing the location of the joint intensity pair. cMI is calculated as the expected value of the cMI between the image intensities given the spatial distribution. The cMI measure was incorporated in a tensor-product B-spline nonrigid registration method, using either a Parzen window or generalized partial volume kernel for histogram construction. cMI was compared to the classical global mutual information (gMI) approach in theoretical, phantom, and clinical settings. We show that cMI significantly outperforms gMI for all applications.

Current ultrasound methods for measuring myocardial strain are often limited to measurements in one or two dimensions. Cardiac motion and deformation however are truly 3-D. With the introduction of matrix transducer technology, 3-D ultrasound imaging of the heart has become feasible but suffers from low temporal and spatial resolution, making 3-D strain estimation challenging. In this paper, it is shown that automatic intensity-based spatio-temporal elastic registration of currently available 3-D volumetric ultrasound data sets can be used to measure the full 3-D strain tensor. The method was validated using simulated 3-D ultrasound data sets of the left ventricle (LV). Three types of data sets were simulated: a normal and symmetric LV with different heart rates, a more realistic asymmetric normal LV and an infarcted LV. The absolute error in the estimated displacement was between 0.47 +/-0.23 and 1.00 +/-0.59 mm, depending on heart rate and amount of background noise. The absolute error on the estimated strain was 9%-21% for the radial strain and 1%-4% for the longitudinal and circumferential strains. No large differences were found between the different types of data sets. The shape of the strain curves was estimated properly and the position of the infarcts could be identified correctly. Preliminary results on clinical data taken in vivo from three healthy volunteers and one patient with an apical aneurism confirmed these findings in a qualitative manner as the strain curves obtained with the proposed method have an amplitude and shape similar to what could be expected.

Multipinhole single photon emission computed tomography (SPECT) imaging has several advantages over single pinhole SPECT imaging, including an increased sensitivity and an improved sampling. However, the quest for a good design is challenging, due to the large number of design parameters. The effect of one of these, the amount of overlap in the projection images, on the reconstruction image quality, is examined in this paper. The evaluation of the quality is based on efficient approximations for the linearized local impulse response and the covariance in a voxel, and on the bias of the reconstruction of the noiseless projection data. Two methods are proposed that remove the overlap in the projection image by blocking certain projection rays with the use of extra shielding between the pinhole plate and the detector. Also two measures to quantify the amount of overlap are suggested. First, the approximate method, predicting the contrast-to-noise ratio (CNR), is validated using postsmoothed maximum likelihood expectation maximization (MLEM) reconstructions with an imposed target resolution. Second, designs with different amounts of overlap are evaluated to study the effect of multiplexing. In addition, the CNR of each pinhole design is also compared with that of the same design where overlap is removed. Third, the results are interpreted with the overlap quantification measures. Fourth, the two proposed overlap removal methods are compared. From the results we can conclude that, once the complete detector area has been used, the extra sensitivity due to multiplexing is only able to compensate for the loss of information, not to improve the CNR. Removing the overlap, however, improves the CNR. The gain is most prominent in the central field of view, though often at the cost of the CNR of some voxels at the edges, since after overlap removal very little information is left for their reconstruction. The reconstruction images provide insight in the multiplexing and truncation artifacts.

During the registration of 3D CAD models of metallic knee implant components to single-plane sagittal fluoroscopy images, the 3D pose of each implant component is estimated by maximizing the similarity between its 2D image appearance and the observed fluoroscopy image. Because knee implant components are highly symmetrical with respect to the sagittal plane, two significantly different model poses result in 2D image projections very similar to an observed sagittal fluoroscopy image. Traditional 2D/3D registration algorithms tend to converge to one of these symmetrical poses discarding the other one. This paper presents a method which simultaneously estimates both symmetrical solutions. In order not to limit the proposed method to 3D models with an exact plane of symmetry, a completely data-driven symmetry constraint is used which is imposed to the estimated pose parameters. The algorithm is embedded in a statistical framework which is optimized using a deterministic annealing expectation-maximization approach. The validity of the method is demonstrated by registration of the tibial knee implant component to real and simulated fluoroscopy images.

A new generic model-based segmentation algorithm is presented, which can be trained from examples akin to the active shape model (ASM) approach in order to acquire knowledge about the shape to be segmented and about the gray-level appearance of the object in the image. Whereas ASM alternates between shape and intensity information during search, the proposed approach optimizes for shape and intensity characteristics simultaneously. Local gray-level appearance information at the landmark points extracted from feature images is used to automatically detect a number of plausible candidate locations for each landmark. The shape information is described by multiple landmark-specific statistical models that capture local dependencies between adjacent landmarks on the shape. The shape and intensity models are combined in a single cost function that is optimized noniteratively using dynamic programming, without the need for initialization. The algorithm was validated for segmentation of anatomical structures in chest and hand radiographs. In each experiment, the presented method had a significant higher performance when compared to the ASM schemes. As the method is highly effective, optimally suited for pathological cases and easy to implement, it is highly useful for many medical image segmentation tasks.

Previously, we developed a method to determine the acquisition geometry of a pinhole camera. This information is needed for the correct reconstruction of pinhole single photon emission computed tomography images. The method uses a calibration phantom consisting of three point sources and their positions in the field of view (FOV) influence the accuracy of the geometry estimate. This paper proposes two particular configurations of point sources with specific positions and orientations in the FOV for optimal image reconstruction accuracy. For the proposed calibration setups, inaccuracies of the geometry estimate due to noise in the calibration data, only cause subresolution inaccuracies in reconstructed images. The calibration method also uses a model of the point source configuration, which is only known with limited accuracy. The study demonstrates, however, that, with the proposed calibration setups, the error in reconstructed images is comparable to the error in the phantom model.

We present a system to assist in the treatment of cardiac arrhythmias by catheter ablation. A patient-specific three-dimensional (3-D) anatomical model, constructed from magnetic resonance images, is merged with fluoroscopic images in an augmented reality environment that enables the transfer of electrocardiography (ECG) measurements and cardiac activation times onto the model. Accurate mapping is realized through the combination of: a new calibration technique, adapted to catheter guided treatments; a visual matching registration technique, allowing the electrophysiologist to align the model with contrast-enhanced images; and the use of virtual catheters, which enable the annotation of multiple ECG measurements on the model. These annotations can be visualized by color coding on the patient model. We provide an accuracy analysis of each of these components independently. Based on simulation and experiments, we determined a segmentation error of 0.6 mm, a calibration error in the order of 1 mm and a target registration error of 1.04 +/- 0.45 mm. The system provides a 3-D visualization of the cardiac activation pattern which may facilitate and improve diagnosis and treatment of the arrhytmia. Because of its low cost and similar advantages we believe our approach can compete with existing commercial solutions, which rely on dedicated hardware and costly catheters. We provide qualitative results of the first clinical use of the system in 11 ablation procedures.

Positron emission tomography (PET) of the cerebral glucose metabolism has shown to be useful in the presurgical evaluation of patients with epilepsy. Between seizures, PET images using fluorodeoxyglucose (FDG) show a decreased glucose metabolism in areas of the gray matter (GM) tissue that are associated with the epileptogenic region. However, detection of subtle hypo-metabolic regions is limited by noise in the projection data and the relatively small thickness of the GM tissue compared to the spatial resolution of the PET system. Therefore, we present an iterative maximum-a-posteriori based reconstruction algorithm, dedicated to the detection of hypo-metabolic regions in FDG-PET images of the brain of epilepsy patients. Anatomical information, derived from magnetic resonance imaging data, and pathophysiological knowledge was included in the reconstruction algorithm. Two Monte Carlo based brain software phantom experiments were used to examine the performance of the algorithm. In the first experiment, we used perfect, and in the second, imperfect anatomical knowledge during the reconstruction process. In both experiments, we measured signal-to-noise ratio (SNR), root mean squared (rms) bias and rms standard deviation. For both experiments, bias was reduced at matched noise levels, when compared to post-smoothed maximum-likelihood expectation-maximization (ML-EM) and maximum a posteriori reconstruction without anatomical priors. The SNR was similar to that of ML-EM with optimal post-smoothing, although the parameters of the prior distributions were not optimized. We can conclude that the use of anatomical information combined with prior information about the underlying pathology is very promising for the detection of subtle hypo-metabolic regions in the brain of patients with epilepsy.

Analysis of multispectral or multitemporal images requires proper geometric alignment of the images to compare corresponding regions in each image volume. Retrospective three-dimensional alignment or registration of multimodal medical images based on features intrinsic to the image data itself is complicated by their different photometric properties, by the complexity of the anatomical objects in the scene and by the large variety of clinical applications in which registration is involved. While the accuracy of registration approaches based on matching of anatomical landmarks or object surfaces suffers from segmentation errors, voxel-based approaches consider all voxels in the image without the need for segmentation. The recent introduction of the criterion of maximization of mutual information, a basic concept from information theory, has proven to be a breakthrough in the field. While solutions for intrapatient affine registration based on this concept are already commercially available, current research in the field focuses on interpatient nonrigid matching.

A method is presented to estimate the acquisition geometry of a pinhole single photon emission computed tomography (SPECT) camera with a circular detector orbit. This information is needed for the reconstruction of tomographic images. The calibration uses the point source projection locations of a tomographic acquisition of three point sources located at known distances from each other. It is shown that this simple phantom provides the necessary and sufficient information for the proposed calibration method. The knowledge of two of the distances between the point sources proves to be essential. The geometry is estimated by fitting analytically calculated projections to the measured ones, using a simple least squares Powell algorithm. Some mild a priori knowledge is used to constrain the solutions of the fit. Several of the geometrical parameters are however highly correlated. The effect of these correlations on the reconstructed images is evaluated in simulation studies and related to the estimation accuracy. The highly correlated detector tilt and electrical shift are shown to be the critical parameters for accurate image reconstruction. The performance of the algorithm is finally demonstrated by phantom measurements. The method is based on a single SPECT scan of a simple calibration phantom, executed immediately after the actual SPECT acquisition. The method is also applicable to cone-beam SPECT and X-ray CT.

Accurate brain tissue segmentation by intensity-based voxel classification of magnetic resonance (MR) images is complicated by partial volume (PV) voxels that contain a mixture of two or more tissue types. In this paper, we present a statistical framework for PV segmentation that encompasses and extends existing techniques. We start from a commonly used parametric statistical image model in which each voxel belongs to one single tissue type, and introduce an additional downsampling step that causes partial voluming along the borders between tissues. An expectation-maximization approach is used to simultaneously estimate the parameters of the resulting model and perform a PV classification. We present results on well-chosen simulated images and on real MR images of the brain, and demonstrate that the use of appropriate spatial prior knowledge not only improves the classifications, but is often indispensable for robust parameter estimation as well. We conclude that general robust PV segmentation of MR brain images requires statistical models that describe the spatial distribution of brain tissues more accurately than currently available models.

We propose a voxel-based nonrigid registration algorithm for temporal subtraction of two-dimensional thorax X-ray computed radiography images of the same subject. The deformation field is represented by a B-spline with a limited number of degrees of freedom, that allows global rib alignment to minimize subtraction artifacts within the lung field without obliterating interval changes of clinically relevant soft-tissue abnormalities. The spline parameters are constrained by a statistical deformation model that is learned from a training set of manually aligned image pairs using principal component analysis. Optimization proceeds along the transformation components rather then along the individual spline coefficients, using pattern intensity of the subtraction image within the automatically segmented lung field region as the criterion to be minimized and applying a simulated annealing strategy for global optimization in the presence of multiple local optima. The impact of different transformation models with varying number of deformation modes is evaluated on a training set of 26 images using a leave-one-out strategy and compared to the manual registration result in terms of criterion value and deformation error. Registration quality is assessed on a second set of validation images by a human expert rating each subtraction image on screen. In 85% of the cases, the registration is subjectively rated to be adequate for clinical use.

Ultrasonic imaging is the noninvasive clinical imaging modality of choice for diagnosing heart disease. At present, two-dimensional ultrasonic grayscale images provide a relatively cheap, fast, bedside method to study the morphology of the heart. Several methods have been proposed to assess myocardial function. These have been based on either grayscale or motion (velocity) information measured in real-time. However, the quantitative assessment of regional myocardial function remains an important goal in clinical cardiology. To do this, ultrasonic strain and strain-rate imaging have been introduced. In the clinical setting, these techniques currently only allow one component of the true three-dimensional deformation to be measured. Clinical, multidimensional strain (rate) information can currently thus only be obtained by combining data acquired using different transducer positions. Nevertheless, given the appropriate postprocessing, the clinical value of these techniques has already been shown. Moreover, multidimensional strain and strain-rate estimation of the heart in vivo by means of a single ultrasound acquisition has been shown to be feasible. In this paper, the new techniques of ultrasonic strain rate and strain imaging of the heart are reviewed in terms of definitions, data acquisition, strain-rate estimation, postprocessing, and parameter extraction. Their clinical validation and relevance will be discussed using clinical examples on relevant cardiac pathology. Based on these examples, suggestions are made for future developments of these techniques.

This paper presents a fully automated algorithm for segmentation of multiple sclerosis (MS) lesions from multispectral magnetic resonance (MR) images. The method performs intensity-based tissue classification using a stochastic model for normal brain images and simultaneously detects MS lesions as outliers that are not well explained by the model. It corrects for MR field inhomogeneities, estimates tissue-specific intensity models from the data itself, and incorporates contextual information in the classification using a Markov random field. The results of the automated method are compared with lesion delineations by human experts, showing a high total lesion load correlation. When the degree of spatial correspondence between segmentations is taken into account, considerable disagreement is found, both between expert segmentations, and between expert and automatic measurements.

A new iterative maximum-likelihood reconstruction algorithm for X-ray computed tomography is presented. The algorithm prevents beam hardening artifacts by incorporating a polychromatic acquisition model. The continuous spectrum of the X-ray tube is modeled as a number of discrete energies. The energy dependence of the attenuation is taken into account by decomposing the linear attenuation coefficient into a photoelectric component and a Compton scatter component. The relative weight of these components is constrained based on prior material assumptions. Excellent results are obtained for simulations and for phantom measurements. Beam-hardening artifacts are effectively eliminated. The relation with existing algorithms is discussed. The results confirm that improving the acquisition model assumed by the reconstruction algorithm results in reduced artifacts. Preliminary results indicate that metal artifact reduction is a very promising application for this new algorithm.

We propose a model-based method for fully automated bias field correction of MR brain images. The MR signal is modeled as a realization of a random process with a parametric probability distribution that is corrupted by a smooth polynomial inhomogeneity or bias field. The method we propose applies an iterative expectation-maximization (EM) strategy that interleaves pixel classification with estimation of class distribution and bias field parameters, improving the likelihood of the model parameters at each iteration. The algorithm, which can handle multichannel data and slice-by-slice constant intensity offsets, is initialized with information from a digital brain atlas about the a priori expected location of tissue classes. This allows full automation of the method without need for user interaction, yielding more objective and reproducible results. We have validated the bias correction algorithm on simulated data and we illustrate its performance on various MR images with important field inhomogeneities. We also relate the proposed algorithm to other bias correction algorithms.

We describe a fully automated method for model-based tissue classification of magnetic resonance (MR) images of the brain. The method interleaves classification with estimation of the model parameters, improving the classification at each iteration. The algorithm is able to segment single- and multispectral MR images, corrects for MR signal inhomogeneities, and incorporates contextual information by means of Markov random Fields (MRF's). A digital brain atlas containing prior expectations about the spatial location of tissue classes is used to initialize the algorithm. This makes the method fully automated and therefore it provides objective and reproducible segmentations. We have validated the technique on simulated as well as on real MR images of the brain.

In order to perform attenuation correction in emission tomography an attenuation map is required. We propose a new method to compute this map directly from the emission sinogram, eliminating the transmission scan from the acquisition protocol. The problem is formulated as an optimization task where the objective function is a combination of the likelihood and an a priori probability. The latter uses a Gibbs prior distribution to encourage local smoothness and a multimodal distribution for the attenuation coefficients. Since the attenuation process is different in positron emission tomography (PET) and single photon emission tomography (SPECT), a separate algorithm for each case is derived. The method has been tested on mathematical phantoms and on a few clinical studies. For PET, good agreement was found between the images obtained with transmission measurements and those produced by the new algorithm in an abdominal study. For SPECT, promising simulation results have been obtained for nonhomogeneous attenuation due to the presence of the lungs.

In recent years, advances in computer technology and a significant increase in the accuracy of medical imaging have made it possible to develop systems that can assist the clinician in diagnosis, planning, and treatment. This paper deals with an area that is generally referred to as computer-assisted surgery, image-directed surgery, or image-guided surgery. We report the research, development, and clinical validation performed since January 1996 in the European Applications in Surgical Interventions (EASI) project, which is funded by the European Commission in their "4th Framework Telematics Applications for Health" program. The goal of this project is the improvement of the effectiveness and quality of image-guided neurosurgery of the brain and image-guided vascular surgery of abdominal aortic aneurysms, while at the same time reducing patient risks and overall cost. We have developed advanced prototype systems for preoperative surgical planning and intraoperative surgical navigation, and we have extensively clinically validated these systems. The prototype systems and the clinical validation results are described in this paper.

A preoperative planning system for oral implant surgery was developed which takes as input computed tomographies (CT's) of the jaws. Two-dimensional (2-D) reslices of these axial CT slices orthogonal to a curve following the jaw arch are computed and shown together with three-dimensional (3-D) surface rendered models of the bone and computer-aided design (CAD)-like implant models. A technique is developed for scanning and visualizing an eventual existing removable prosthesis together with the bone structures. Evaluation of the planning done with the system shows a difference between 2-D and 3-D planning methods. Validation studies measure the benefits of the 3-D approach by comparing plans made in 2-D mode only with those further adjusted using the full 3-D visualization capabilities of the system. The benefits of a 3-D approach are then evident where a prosthesis is involved in the planning. For the majority of the patients, clinically important adjustments and optimizations to the 2-D plans are made once the 3-D visualization is enabled, effectively resulting in a better plan. The alterations are related to bone quality and quantity (p < 0.05), biomechanics (p < 0.005), and esthetics (p < 0.005), and are so obvious that the 3-D plan stands out clearly (p < 0.005). The improvements often avoid complications such as mandibular nerve damage, sinus perforations, fenestrations, or dehiscences.

A new approach to the problem of multimodality medical image registration is proposed, using a basic concept from information theory, mutual information (MI), or relative entropy, as a new matching criterion. The method presented in this paper applies MI to measure the statistical dependence or information redundancy between the image intensities of corresponding voxels in both images, which is assumed to be maximal if the images are geometrically aligned. Maximization of MI is a very general and powerful criterion, because no assumptions are made regarding the nature of this dependence and no limiting constraints are imposed on the image content of the modalities involved. The accuracy of the MI criterion is validated for rigid body registration of computed tomography (CT), magnetic resonance (MR), and photon emission tomography (PET) images by comparison with the stereotactic registration solution, while robustness is evaluated with respect to implementation issues, such as interpolation and optimization, and image content, including partial overlap and image degradation. Our results demonstrate that subvoxel accuracy with respect to the stereotactic reference solution can be achieved completely automatically and without any prior segmentation, feature extraction, or other preprocessing steps which makes this method very well suited for clinical applications.

In X-ray projection radiography, especially if area detectors are used, scattered radiation can strongly degrade the image contrast. Although the amount of radiation scatter is partly reduced by the use of anti-scatter grids or air gaps, the contrast is mostly still substantially degraded in large areas of the images. Here, the radiation scatter in radiographs is modeled as the sum of a low-pass filtered version of the primary radiation distribution and a residual scatter component. The authors show how the Multiscale-Image-Contrast-Amplification (MUSICA) algorithm for contrast enhancement can automatically remove the degrading effect of scattered radiation on image contrast, by selectively suppressing low resolution components, without significantly affecting the primary radiation distribution at any spatial resolution.

The authors' work contributes to the accurate segmentation of blood vessels from 3D medical images. The blood vessel axis and surface are optimized in an alternating way. Starting from an initial blood vessel axis estimate, slices are resampled in the 3D data volume perpendicular to this axis. In these slices, blood vessel contour candidate points are extracted at maximum gradient positions on a star pattern. The selection of a closed contour among these candidates is optimized with respect to a cost function by dynamic programming. The blood vessel axis is re-estimated at the center of the extracted contours and the process is repeated until convergence. Results are shown on both synthetic and real spiral CT angiographic images.

The objective of this study is to establish a protocol for the technical and clinical evaluation of a workstation for the planning of stereotactic neurosurgical interventions that has been developed in the framework of a joint European research project. Although several such workstations have been proposed before, they lacked the final and most important step, that of clinical validation. They failed to rigorously prove that their product was useful. The authors present a new method that is applicable to the evaluation of a wide range of medical technologies. Their protocol basically assesses the clinical relevance of the user requirements that are at the root of the development of the new technology. The evaluation consists of two stages. During functional specification, iterative prototyping is used to establish the clinical requirements and to assure the quality of the final product. A case study design is used in a second stage that assesses the clinical usability. A before-after study gives a first indication of cost effectiveness and improvement of health care quality.

Accurate reconstruction of SPECT images is hampered by four nonlinear effects in the acquisition process: attenuation, scatter, collimator acceptance angle, and statistical noise. A good mathematical description of these effects is obviously crucial for the reconstruction. Poisson noise, attenuation, and collimator acceptance angle are relatively easy to model. Scattering, however, is a very complex process, and is mostly described using empirical models. A new model for the scatter point spread function in a homogeneous medium that is based on physical considerations and is capable of predicting scatter contributions of point sources as a function of depth in homogeneous media is presented. The model is verified with Monte Carlo simulations and line source measurements, and is compared to an existing empirical model.

Algorithms that calculate maximum likelihood (ML) and maximum a posteriori solutions using expectation-maximization have been successfully applied to SPECT and PET. These algorithms are appealing because of their solid theoretical basis and their guaranteed convergence. A major drawback is the slow convergence, which results in-long processing times. The authors present 2 new heuristic acceleration methods for maximum likelihood reconstruction of ECT images. The first method incorporates a frequency-dependent amplification in the calculations, to compensate for the low pass filtering of the backprojection operation. In the second method, an amplification factor is incorporated that suppresses the effect of attenuation on the updating factors. Both methods are compared to the 1-dimensional line search method proposed by Lewitt. All 3 methods accelerate the ML algorithm. On the authors' test images, Lewitt's method produced the strongest acceleration of the three individual methods. However, the combination of the frequency amplification with the line search method results in a new algorithm with still better performance. Under certain conditions, an effective frequency amplification can be already achieved by skipping some of the calculations required for ML.

The recognition of the coronary blood vessels on single angiograms is addressed. Recognizing blood vessels on digital subtraction angiograms is a prerequisite to automatic diagnosis and quantification of blood-vessel pathologies. It is shown how stereoscopic angiograms can improve the interpretation process. The computational strategy recommended to solve the problem has the following important characteristics. First, the model is hierarchically instantiated, starting from a partial model and working up to the final complete model. Second, the contact with the image is never broken during the interpretation process. Missing model attributes are instantiated by returning to the image. Third, delineation (segmentation and image partitioning), interpretation, and stereoscopic matching are not sequential but integrated processes. Fourth, the knowledge representation is partially iconic. The iconic memory is used to guide the search for missing or incomplete attributes in the image.>

Expert systems and image understanding have traditionally been considered as two separate application fields of artificial intelligence (AI). In this paper it is shown, however, that the idea of building an expert system for image understanding may be fruitful. Although this paper may serve as a framework for situating existing works on knowledge-based vision, it is not a review paper. The interested reader will therefore be referred to some recommended survey papers in the literature.