Medical Image Analysis

Editorial board

2012-05-01

Cardiac MR perfusion image processing techniques: A survey

2012-01-11

Graphical abstract: Highlights: ► Survey of semi- and fully automatic image processing methods for cardiac MR perfusion quantification. ► Classification of image processing methods based on their registration, segmentation and multimodality fusion and visualization algorithms. ► Investigation of the advantages and drawbacks of the surveyed methods and their robustness to acquisition artifacts. ► Extensive literature overview of the last 20 years.Abstract: First-pass cardiac MR perfusion (CMRP) imaging has undergone rapid technical advancements in recent years. Although the efficacy of CMRP imaging in the assessment of coronary artery diseases (CAD) has been proven, its clinical use is still limited. This limitation stems, in part, from manual interaction required to quantitatively analyze the large amount of data. This process is tedious, time-consuming, and prone to operator bias. Furthermore, acquisition and patient related image artifacts reduce the accuracy of quantitative perfusion assessment. With the advent of semi- and fully automatic image processing methods, not only the challenges posed by these artifacts have been overcome to a large extent, but a significant reduction has also been achieved in analysis time and operator bias. Despite an extensive literature on such image processing methods, to date, no survey has been performed to discuss this dynamic field. The purpose of this article is to provide an overview of the current state of the field with a categorical study, along with a future perspective on the clinical acceptance of image processing methods in the diagnosis of CAD.

Mass preserving image registration for lung CT

2012-01-16

Graphical abstract: Improvement of mass preserving image registration for four datasets with increasing differences in lung volume.Highlights: ► Preservation of mass is a plausible model of lung parenchyma. ► Mass preserving model simulates lung tissue density change related to the change in regional ventilation. ► Incorporating mass preserving model into an image registration leads to more accurate results.Abstract: This paper presents a mass preserving image registration algorithm for lung CT images. To account for the local change in lung tissue intensity during the breathing cycle, a tissue appearance model based on the principle of preservation of total lung mass is proposed. This model is incorporated into a standard image registration framework with a composition of a global affine and several free-form B-Spline transformations with increasing grid resolution. The proposed mass preserving registration method is compared to registration using the sum of squared intensity differences as a similarity function on four groups of data: 44 pairs of longitudinal inspiratory chest CT scans with small difference in lung volume; 44 pairs of longitudinal inspiratory chest CT scans with large difference in lung volume; 16 pairs of expiratory and inspiratory CT scans; and 5 pairs of images extracted at end exhale and end inhale phases of 4D-CT images. Registration errors, measured as the average distance between vessel tree centerlines in the matched images, are significantly lower for the proposed mass preserving image registration method in the second, third and fourth group, while there is no statistically significant difference between the two methods in the first group. Target registration error, assessed via a set of manually annotated landmarks in the last group, was significantly smaller for the proposed registration method.

3D segmentation of SBFSEM images of neuropil by a graphical model over supervoxel boundaries

2011-12-21

Graphical abstract: Highlight: ► We propose a supervoxel merging method for segmenting 3D SBFSEM images of neuropil. ► All merging decisions are coupled in a graphical model. ► Knowledge about correct segments is incorporated in higher-order potentials. ► These potentials are learned automatically from training data. ► The scope of this improvement is demonstrated on the benchmark dataset E1088.Abstract: The segmentation of large volume images of neuropil acquired by serial sectioning electron microscopy is an important step toward the 3D reconstruction of neural circuits. The only cue provided by the data at hand is boundaries between otherwise indistinguishable objects. This indistinguishability, combined with the boundaries becoming very thin or faint in places, makes the large body of work on region-based segmentation methods inapplicable. On the other hand, boundary-based methods that exploit purely local evidence do not reach the extremely high accuracy required by the application domain that cannot tolerate the global topological errors arising from false local decisions. As a consequence, we propose a supervoxel merging method that arrives at its decisions in a non-local fashion, by posing and approximately solving a joint combinatorial optimization problem over all faces between supervoxels. The use of supervoxels allows the extraction of expressive geometric features. These are used by the higher-order potentials in a graphical model that assimilate knowledge about the geometry of neural surfaces by automated training on a gold standard. The scope of this improvement is demonstrated on the benchmark dataset E1088 () of 7.5billionvoxels from the inner plexiform layer of rabbit retina. We provide C++ source code for annotation, geometry extraction, training and inference.

Manifold learning for image-based breathing gating in ultrasound and MRI

Christian Wachinger, Mehmet Yigitsoy, Erik-Jan Rijkhorst, Nassir Navab — 2011-12-09

Graphical abstract: Highlights: ► Automatic, image-based gating in ultrasound and MRI. ► Application of Laplacian eigenmaps for gating. ► Creation of 4D ultrasound data with wobbler transducer. ► Validation by comparison to alternative gating approaches.Abstract: Respiratory motion is a challenging factor for image acquisition and image-guided procedures in the abdominal and thoracic region. In order to address the issues arising from respiratory motion, it is often necessary to detect the respiratory signal. In this article, we propose a novel, purely image-based retrospective respiratory gating method for ultrasound and MRI. Further, we apply this technique to acquire breathing-affected 4D ultrasound with a wobbler probe and, similarly, to create 4D MR with a slice stacking approach. We achieve the gating with Laplacian eigenmaps, a manifold learning technique, to determine the low-dimensional manifold embedded in the high-dimensional image space. Since Laplacian eigenmaps assign to each image frame a coordinate in low-dimensional space by respecting the neighborhood relationship, they are well suited for analyzing the breathing cycle. We perform the image-based gating on several 2D and 3D ultrasound datasets over time, and quantify its very good performance by comparing it to measurements from an external gating system. For MRI, we perform the manifold learning on several datasets for various orientations and positions. We achieve very high correlations by a comparison to an alternative gating with diaphragm tracking.

Nonlinear dimensionality reduction combining MR imaging with non-imaging information

2011-12-26

Graphical abstract: The top figure shows a 2D manifold embedding of images acquired from subjects with AD (squares) and healthy controls (circles) using pairwise image similarities with Laplacian Eigenmaps. The bottom figure shows an embedding obtained with the proposed method incorporating metadata (color-coded) into the manifold learning process. A better separation between the two groups can be obtained when considering both measures. Misclassified subjects are marked by a black outline.Highlights: ► We propose an extension to nonlinear dimensionality reduction with Laplacian Eigenmaps. ► MR imaging information is combined with clinical metadata. ► The representation of subjects in the low-dimensional space is interpreted as biomarker. ► We use this biomarker to classify AD subjects from healthy controls. ► We evaluate the method on the ADNI study.Abstract: We propose a framework for the extraction of biomarkers from low-dimensional manifolds representing inter-subject brain variation. Manifold coordinates of each image capture information about structural shape and appearance and, when a phenotype exists, about the subject’s clinical state. Our framework incorporates subject meta-information into the manifold learning step. Apart from gender and age, information such as genotype or a derived biomarker is often available in clinical studies and can inform the classification of a query subject. Such information, whether discrete or continuous, is used as an additional input to manifold learning, extending the Laplacian Eigenmap objective function and enriching a similarity measure derived from pairwise image similarities. The biomarkers identified with the proposed method are data-driven in contrast to a priori defined biomarkers derived from, e.g., manual or automated segmentations. They form a unified representation of both the imaging and non-imaging measurements, providing a natural use for data analysis and visualization. We test the method to classify subjects with Alzheimer’s Disease (AD), mild cognitive impairment (MCI) and healthy controls enrolled in the ADNI study. Non-imaging metadata used are ApoE genotype, a risk factor associated with AD, and the CSF-concentration of , an established biomarker for AD. In addition, we use hippocampal volume as a derived imaging-biomarker to enrich the learned manifold. Our classification results compare favorably to what has been reported in a recent meta-analysis using established neuroimaging methods on the same database.

A fast analysis method for non-invasive imaging of blood flow in individual cerebral arteries using vessel-encoded arterial spin labelling angiography

2011-12-29

Graphical abstract: Highlights: ► Non-invasive vessel selective MR angiography. ► Fast Bayesian analysis for artery flow contributions. ► Robust treatment of imperfect artery location specification e.g. due to patient movement. ► Robust to clinical scenarios, e.g. occluded arteries.Abstract: Arterial spin labelling (ASL) MRI offers a non-invasive means to create blood-borne contrast in vivo for dynamic angiographic imaging. By spatial modulation of the ASL process it is possible to uniquely label individual arteries over a series of measurements, allowing each to be separately identified in the resulting angiographic images. This separation requires appropriate analysis for which a general Bayesian framework has previously been proposed. Here this framework is adapted for clinical dynamic angiographic imaging. This specifically addresses the issues of computational speed of the algorithm and the robustness required to deal with real patient data. An algorithm is proposed that can incorporate planning information about the arteries being imaged whilst adapting for subsequent patient movement. A fast maximum a posteriori solution is adopted and shown to be only marginally less accurate than Monte Carlo sampling under simulation. The final algorithm is demonstrated on in vivo data with analysis on a time scale of the order of 10min, from both a healthy control and a patient with a vertebro-basilar occlusion.

Brain tissue segmentation in MR images based on a hybrid of MRF and social algorithms

Sahar Yousefi, Reza Azmi, Morteza Zahedi — 2012-02-02

Graphical abstract: Effective abnormality detection and diagnosis in Magnetic Resonance Images (MRIs) requires a robust segmentation strategy. Since manual segmentation is a so time-consuming task which engages valuable human resources, automatic MRI segmentation have been received an enormous amount of attention. For this purpose various techniques have been applied however Markov Random Field (MRF) based algorithms have produced better results in noisy images compared to other methods. MRF seeks for a label field which minimizes energy function. Traditional minimization method, simulated annealing (SA), uses Monte Carlo simulation to access the minimum solution with heavy computation burden. For this reason MRFs never be used in real time processing environments. This paper proposed a novel method based on MRF and hybrid of social algorithms contains ant colony optimization (ACO) and Gossiping algorithm. Combining ACO with Gossiping algorithm assists ants to find the better path using the information of their neighbors. Therefore, this interaction causes the algorithm converges to optimum solution faster. Several experiments on phantom and real images were performed. Results indicate the proposed algorithm outperforms the traditional MRF in speed and accuracy.Highlights: ► This article presents a novel unsupervised MRF-based model for MRI segmentation. ► The proposed method uses ACO and gossiping algorithm to speed up the classic MRF. ► Tackling Gossiping algorithm, the proposed method assists ants in smart decision. ► The IBSR and Brainweb datasets have been used for testing the method. ► Results demonstrate the novel method outperforms other models in speed and quality.Abstract: Effective abnormality detection and diagnosis in Magnetic Resonance Images (MRIs) requires a robust segmentation strategy. Since manual segmentation is a time-consuming task which engages valuable human resources, automatic MRI segmentations received an enormous amount of attention. For this goal, various techniques have been applied. However, Markov Random Field (MRF) based algorithms have produced reasonable results in noisy images compared to other methods. MRF seeks a label field which minimizes an energy function. The traditional minimization method, simulated annealing (SA), uses Monte Carlo simulation to access the minimum solution with heavy computation burden. For this reason, MRFs are rarely used in real time processing environments. This paper proposed a novel method based on MRF and a hybrid of social algorithms that contain an ant colony optimization (ACO) and a Gossiping algorithm which can be used for segmenting single and multispectral MRIs in real time environments. Combining ACO with the Gossiping algorithm helps find the better path using neighborhood information. Therefore, this interaction causes the algorithm to converge to an optimum solution faster. Several experiments on phantom and real images were performed. Results indicate that the proposed algorithm outperforms the traditional MRF and hybrid of MRF-ACO in speed and accuracy.

A CANDLE for a deeper in vivo insight

2012-01-20

Graphical abstract: Optic chiasm between 275μm and 320μmHighlights: ► A new method for the denoising of confocal and multiphoton images is proposed. ► An extensive validation demonstrates the high denoising performance of our filter. ► The deeper in vivo imaging capabilities enabled by our filter are presented.Abstract: A new Collaborative Approach for eNhanced Denoising under Low-light Excitation (CANDLE) is introduced for the processing of 3D laser scanning multiphoton microscopy images. CANDLE is designed to be robust for low signal-to-noise ratio (SNR) conditions typically encountered when imaging deep in scattering biological specimens. Based on an optimized non-local means filter involving the comparison of filtered patches, CANDLE locally adapts the amount of smoothing in order to deal with the noise inhomogeneity inherent to laser scanning fluorescence microscopy images. An extensive validation on synthetic data, images acquired on microspheres and in vivo images is presented. These experiments show that the CANDLE filter obtained competitive results compared to a state-of-the-art method and a locally adaptive optimized non-local means filter, especially under low SNR conditions (PSNR<8dB). Finally, the deeper imaging capabilities enabled by the proposed filter are demonstrated on deep tissue in vivo images of neurons and fine axonal processes in the Xenopus tadpole brain.

A probabilistic framework for image information fusion with an application to mammographic analysis

Marina Velikova, Peter J.F. Lucas, Maurice Samulski, Nico Karssemeijer — 2012-01-27

Graphical abstract: Highlights: ► We develop a multi-stage probabilistic framework for multi-view image analysis. ► The framework is computationally efficient and quite general to apply to any domain. ► A sound unbiased evaluation procedure for multi-stage models is proposed. ► Comparison with previous mammographic CAD systems show improved performance.Abstract: The recent increased interest in information fusion methods for solving complex problem, such as in image analysis, is motivated by the wish to better exploit the multitude of information, available from different sources, to enhance decision-making. In this paper, we propose a novel method, that advances the state of the art of fusing image information from different views, based on a special class of probabilistic graphical models, called causal independence models. The strength of this method is its ability to systematically and naturally capture uncertain domain knowledge, while performing information fusion in a computationally efficient way. We examine the value of the method for mammographic analysis and demonstrate its advantages in terms of explicit knowledge representation and accuracy (increase of at least 6.3% and 5.2% of true positive detection rates at 5% and 10% false positive rates) in comparison with previous single-view and multi-view systems, and benchmark fusion methods such as naïve Bayes and logistic regression.

Nonparametric Bayesian inference of the fiber orientation distribution from diffusion-weighted MR images

Enrico Kaden, Frithjof Kruggel — 2012-02-02

Graphical abstract: Bayesian density estimation allows us to disentangle three crossing fiber bundles in the centrum semiovale.Highlights: ► The spherical convolution model is reformulated by adopting a measure-theoretic approach. ► The fiber orientation density is represented by a Dirichlet process mixture, which may be used for realistic simulations. ► The inverse problem is solved by nonparametric Bayesian statistics under a Rician noise model. ► The density estimation framework is demonstrated with HARDI datasets.Abstract: Diffusion MR imaging provides a unique tool to probe the microgeometry of nervous tissue and to explore the wiring diagram of the neural connections noninvasively. Generally, a forward model is established to map the intra-voxel fiber architecture onto the observable diffusion signals, which is reformulated in this article by adopting a measure-theoretic approach. However, the inverse problem, i.e., the spherical deconvolution of the fiber orientation density from noisy MR measurements, is ill-posed. We propose a nonparametric representation of the tangential distribution of the nerve fibers in terms of a Dirichlet process mixture. Given a second-order approximation of the impulse response of a fiber segment, the specified problem is solved by Bayesian statistics under a Rician noise model, using an adaptive reversible jump Markov chain Monte Carlo sampler. The density estimation framework is demonstrated by various experiments with a diffusion MR dataset featuring high angular resolution, uncovering the fiber orientation field in the cerebral white matter of the living human brain.

Automated landmarking and geometric characterization of the carotid siphon

2012-02-10

Graphical abstract: Highlights: ► Automated identification of internal carotid artery. ► Automated detection of vessel bends and anatomical landmarking of carotid siphon. ► Computation of geometric quantities known to have influence on vascular hemodynamics. ► The use of LDDMCM for vessels and the importance of using landmark matching.Abstract: The geometry of the carotid siphon has a large variability between subjects, which has prompted its study as a potential geometric risk factor for the onset of vascular pathologies on and off the internal carotid artery (ICA). In this work, we present a methodology for an objective and extensive geometric characterization of carotid siphon parameterized by a set of anatomical landmarks. We introduce a complete and automated characterization pipeline. Starting from the segmentation of vasculature from angiographic image and its centerline extraction, we first identify ICA by characterizing vessel tree bifurcations and training a support vector machine classifier to detect ICA terminal bifurcation. On ICA centerline curve, we detect anatomical landmarks of carotid siphon by modeling it as a sequence of four bends and selecting their centers and interfaces between them. Bends are detected from the trajectory of the curvature vector expressed in the parallel transport frame of the curve. Finally, using the detected landmarks, we characterize the geometry in two complementary ways. First, with a set of local and global geometric features, known to affect hemodynamics. Second, using large deformation diffeomorphic metric curve mapping (LDDMCM) to quantify pairwise shape similarity. We processed 96 images acquired with 3D rotational angiography. ICA identification had a cross-validation success rate of 99%. Automated landmarking was validated by computing limits of agreement with the reference taken to be the locations of the manually placed landmarks averaged across multiple observers. For all but one landmark, either the bias was not statistically significant or the variability was within 50% of the inter-observer one. The subsequently computed values of geometric features and LDDMCM were commensurate to the ones obtained with manual landmarking. The characterization based on pair-wise LDDMCM proved better in classifying the carotid siphon shape classes than the one based on geometric features. The proposed characterization provides a rich description of geometry and is ready to be applied in the search for geometric risk factors of the carotid siphon.

Statistical 4D graphs for multi-organ abdominal segmentation from multiphase CT

Marius George Linguraru, John A. Pura, Vivek Pamulapati, Ronald M. Summers — 2012-02-13

Graphical abstract: Highlight: ► Anatomical and physiological models were incorporated in statistical graph cuts. ► Multiple abdominal organs were automatically segmented and analyzed. ► Appearance, shape and location priors improved the accuracy of organ segmentation. ► Livers, spleens and kidneys were segmented with volume overlaps over 93.6%..Abstract: The interpretation of medical images benefits from anatomical and physiological priors to optimize computer-aided diagnosis applications. Diagnosis also relies on the comprehensive analysis of multiple organs and quantitative measures of soft tissue. An automated method optimized for medical image data is presented for the simultaneous segmentation of four abdominal organs from 4D CT data using graph cuts. Contrast-enhanced CT scans were obtained at two phases: non-contrast and portal venous. Intra-patient data were spatially normalized by non-linear registration. Then 4D convolution using population training information of contrast-enhanced liver, spleen and kidneys was applied to multiphase data to initialize the 4D graph and adapt to patient-specific data. CT enhancement information and constraints on shape, from Parzen windows, and location, from a probabilistic atlas, were input into a new formulation of a 4D graph. Comparative results demonstrate the effects of appearance, enhancement, shape and location on organ segmentation. All four abdominal organs were segmented robustly and accurately with volume overlaps over 93.6% and average surface distances below 1.1mm.

Image-based characterization of thrombus formation in time-lapse DIC microscopy

2012-02-13

Graphical abstract: Highlights: ► Automatized characterization of thrombus formation in time-lapse microscopy. ► Novel energy model for segmentation of multiple dynamic regions. ► Novel algorithm for the joint segmentation of thrombus and aortic regions. ► Exhaustive validation on synthetic and real microscopic data.Abstract: The characterization of thrombus formation in time-lapse DIC microscopy is of increased interest for identifying genes which account for atherothrombosis and coronary artery diseases (CADs). In particular, we are interested in large-scale studies on zebrafish, which result in large amount of data, and require automatic processing. In this work, we present an image-based solution for the automatized extraction of parameters quantifying the temporal development of thrombotic plugs. Our system is based on the joint segmentation of thrombotic and aortic regions over time. This task is made difficult by the low contrast and the high dynamic conditions observed in vivo DIC microscopic scenes. Our key idea is to perform this segmentation by distinguishing the different motion patterns in image time series rather than by solving standard image segmentation tasks in each image frame. Thus, we are able to compensate for the poor imaging conditions. We model motion patterns by energies based on the idea of dynamic textures, and regularize the model by two prior energies on the shape of the aortic region and on the topological relationship between the thrombus and the aorta. We demonstrate the performance of our segmentation algorithm by qualitative and quantitative experiments on synthetic examples as well as on real in vivo microscopic sequences.