Amir Alansary

MRBrainS challenge: online evaluation framework for brain image segmentation in 3T MRI scans

Many methods have been proposed for tissue segmentation in brain MRI scans. The multitude of methods proposed complicates the choice of one method above others. We have therefore established the MRBrainS online evaluation framework for evaluating (semi)automatic algorithms that segment gray matter (GM), white matter (WM), and cerebrospinal fluid (CSF) on 3T brain MRI scans of elderly subjects (65–80 y). Participants apply their algorithms to the provided data, after which their results are evaluated and ranked. Full manual segmentations of GM, WM, and CSF are available for all scans and used as the reference standard. Five datasets are provided for training and fifteen for testing. The evaluated methods are ranked based on their overall performance to segment GM, WM, and CSF and evaluated using three evaluation metrics (Dice, H95, and AVD) and the results are published on the MRBrainS13 website. We present the results of eleven segmentation algorithms that participated in the MRBrainS13 challenge workshop at MICCAI, where the framework was launched, and three commonly used freeware packages: FreeSurfer, FSL, and SPM. The MRBrainS evaluation framework provides an objective and direct comparison of all evaluated algorithms and can aid in selecting the best performing method for the segmentation goal at hand.

Segmentation of lung region based on using parallel implementation of joint MGRF: Validation on 3D realistic lung phantoms

The segmentation of the lung tissues in chest Computed Tomography (CT) images is an important step for developing any Computer-Aided Diagnostic (CAD) system for lung cancer and other pulmonary diseases. In this paper, we introduce a new framework to generate 3D realistic synthetic phantoms to validate our developed Joint Markov-Gibbs based lung segmentation approach from CT data. Our framework is based on using a 3D generalized Gauss-Markov Random Field (GGMRF) model of voxel intensities with pairwise interaction to model the 3D appearance of the lung tissues. Then, the appearance of the generated 3D phantoms is simulated based on iterative minimization of an energy function that is based on using the learned 3D-GGMRF image model. These 3D realistic phantoms can be used to evaluate the performance of any lung segmentation approach. In this paper, we used the 3D realistic phantoms to evaluate the performance of our developed lung segmentation approach based on using the Dice Similarity Coefficient (DSC) metric and the Receiver Operating Characteristics (ROC). The DSC demonstrated that our approach achieves a mean DSC value of 0.994 ± 0.0034. Moreover, the ROC analysis for our method showed the best performance (area 0.99), while intensity showed the worst performance (area 0.92).

Flexible Reconstruction and Correction of Unpredictable Motion from Stacks of 2D Images

We present a method to correct motion in fetal in-utero scan sequences. The proposed approach avoids previously necessary manual segmentation of a region of interest. We solve the problem of non-rigid motion by splitting motion corrupted slices into overlapping patches of finite size. In these patches the assumption of rigid motion approximately holds and they can thus be used to perform a slice-to-volume-based (SVR) reconstruction during which their consistency with the other patches is learned. The learned information is used to reject patches that are not conform with the motion corrected reconstruction in their local areas. We evaluate rectangular and evenly distributed patches for the reconstruction as well as patches that have been derived from super-pixels. Both approaches achieve on 29 subjects aged between 22–37 weeks a sufficient reconstruction quality and facilitate following 3D segmentation of fetal organs and the placenta.

Computer-Aided Diagnosis Systems for Acute Renal Transplant Rejection: Challenges and Methodologies

This chapter overviews one of the most critical problems in urology, namely detection of acute renal transplant rejection. Developing an effective, fast, and accurate computer-aided diagnosis (CAD) system for early detection of acute renal rejection is of great clinical importance for the management of these patients. For this reason, CAD systems for early detection of renal transplant rejection have been investigated in a huge number of research studies using different image modalities, such as ultrasound (US), magnetic resonance imaging (MRI), computed tomography (CT), and radionuclide imaging. A typical CAD system for kidney diagnosis consists of a set of processing steps including, but not limited to, image registration to account for kidney motion, segmentation of the kidney and/or its compartments (e.g., cortex, medulla), construction of agent kinetic curves, functional parameters estimation, and diagnosis and assessment of the kidney status. Due to the widespread popularity of US and MRI, this chapter overviews the current state-of-the-art CAD systems that have been developed for kidney diagnosis using these two image modalities. In addition, the chapter addresses several challenges that researchers face in developing efficient, fast, and reliable CAD systems for early detection of kidney diseases.

Infant Brain Extraction in T1-Weighted MR Images Using BET and Refinement Using LCDG and MGRF Models

In this paper, we propose a novel framework for the automated extraction of the brain from T1-weighted MR images. The proposed approach is primarily based on the integration of a stochastic model [a two-level Markov-Gibbs random field (MGRF)] that serves to learn the visual appearance of the brain texture, and a geometric model (the brain isosurfaces) that preserves the brain geometry during the extraction process. The proposed framework consists of three main steps: 1) Following bias correction of the brain, a new three-dimensional (3-D) MGRF having a 26-pairwise interaction model is applied to enhance the homogeneity of MR images and preserve the 3-D edges between different brain tissues. 2) The nonbrain tissue found in the MR images is initially removed using the brain extraction tool (BET), and then the brain is parceled to nested isosurfaces using a fast marching level set method. 3) Finally, a classification step is applied in order to accurately remove the remaining parts of the skull without distorting the brain geometry. The classification of each voxel found on the isosurfaces is made based on the first- and second-order visual appearance features. The first-order visual appearance is estimated using a linear combination of discrete Gaussians (LCDG) to model the intensity distribution of the brain signals. The second-order visual appearance is constructed using an MGRF model with analytically estimated parameters. The fusion of the LCDG and MGRF, along with their analytical estimation, allows the approach to be fast and accurate for use in clinical applications. The proposed approach was tested on in vivo data using 300 infant 3-D MR brain scans, which were qualitatively validated by an MR expert. In addition, it was quantitatively validated using 30 datasets based on three metrics: the Dice coefficient, the 95% modified Hausdorff distance, and absolute brain volume difference. Results showed the capability of the proposed approach, outperforming four widely used BETs: BET, BET2, brain surface extractor, and infant brain extraction and analysis toolbox. Experiments conducted also proved that the proposed framework can be generalized to adult brain extraction as well.

Fast Fully Automatic Segmentation of the Human Placenta from Motion Corrupted MRI

Recently, magnetic resonance imaging has revealed to be important for the evaluation of placenta’s health during pregnancy. Quantitative assessment of the placenta requires a segmentation, which proves to be challenging because of the high variability of its position, orientation, shape and appearance. Moreover, image acquisition is corrupted by motion artifacts from both fetal and maternal movements. In this paper we propose a fully automatic segmentation framework of the placenta from structural T2-weighted scans of the whole uterus, as well as an extension in order to provide an intuitive pre-natal view into this vital organ. We adopt a 3D multi-scale convolutional neural network to automatically identify placental candidate pixels. The resulting classification is subsequently refined by a 3D dense conditional random field, so that a high resolution placental volume can be reconstructed from multiple overlapping stacks of slices. Our segmentation framework has been tested on 66 subjects at gestational ages 20–38 weeks achieving a Dice score of \(71.95\pm 19.79\,\%\) for healthy fetuses with a fixed scan sequence and \(66.89\pm 15.35\,\%\) for a cohort mixed with cases of intrauterine fetal growth restriction using varying scan parameters.

Dynamic MRI-based computer aided diagnostic systems for early detection of kidney transplant rejection: A survey

Early detection of renal transplant rejection is important to implement appropriate medical and immune therapy in patients with transplanted kidneys. In literature, a large number of computer-aided diagnostic (CAD) systems using different image modalities, such as ultrasound (US), magnetic resonance imaging (MRI), computed tomography (CT), and radionuclide imaging, have been proposed for early detection of kidney diseases. A typical CAD system for kidney diagnosis consists of a set of processing steps including: motion correction, segmentation of the kidney and/or its internal structures (e.g., cortex, medulla), construction of agent kinetic curves, functional parameter estimation, diagnosis, and assessment of the kidney status. In this paper, we survey the current state-of-the-art CAD systems that have been developed for kidney disease diagnosis using dynamic MRI. In addition, the paper addresses several challenges that researchers face in developing efficient, fast and reliable CAD systems for the early...

Predicting slice-to-volume transformation in presence of arbitrary subject motion

This paper aims to solve a fundamental problem in intensity-based 2D/3D registration, which concerns the limited capture range and need for very good initialization of state-of-the-art image registration methods. We propose a regression approach that learns to predict rotations and translations of arbitrary 2D image slices from 3D volumes, with respect to a learned canonical atlas co-ordinate system. To this end, we utilize Convolutional Neural Networks (CNNs) to learn the highly complex regression function that maps 2D image slices into their correct position and orientation in 3D space. Our approach is attractive in challenging imaging scenarios, where significant subject motion complicates reconstruction performance of 3D volumes from 2D slice data. We extensively evaluate the effectiveness of our approach quantitatively on simulated MRI brain data with extreme random motion. We further demonstrate qualitative results on fetal MRI where our method is integrated into a full reconstruction and motion compensation pipeline. With our CNN regression approach we obtain an average prediction error of 7 mm on simulated data, and convincing reconstruction quality of images of very young fetuses where previous methods fail. We further discuss applications to Computed Tomography (CT) and X-Ray projections. Our approach is a general solution to the 2D/3D initialization problem. It is computationally efficient, with prediction times per slice of a few milliseconds, making it suitable for real-time scenarios.

PVR: Patch-to-Volume Reconstruction for Large Area Motion Correction of Fetal MRI

In this paper, we present a novel method for the correction of motion artifacts that are present in fetal magnetic resonance imaging (MRI) scans of the whole uterus. Contrary to current slice-to-volume registration (SVR) methods, requiring an inflexible anatomical enclosure of a single investigated organ, the proposed patch-to-volume reconstruction (PVR) approach is able to reconstruct a large field of view of non-rigidly deforming structures. It relaxes rigid motion assumptions by introducing a specific amount of redundant information that is exploited with parallelized patchwise optimization, super-resolution, and automatic outlier rejection. We further describe and provide an efficient parallel implementation of PVR allowing its execution within reasonable time on commercially available graphics processing units, enabling its use in the clinical practice. We evaluate PVR’s computational overhead compared with standard methods and observe improved reconstruction accuracy in the presence of affine motion artifacts compared with conventional SVR in synthetic experiments. Furthermore, we have evaluated our method qualitatively and quantitatively on real fetal MRI data subject to maternal breathing and sudden fetal movements. We evaluate peak-signal-to-noise ratio, structural similarity index, and cross correlation with respect to the originally acquired data and provide a method for visual inspection of reconstruction uncertainty. We further evaluate the distance error for selected anatomical landmarks in the fetal head, as well as calculating the mean and maximum displacements resulting from automatic non-rigid registration to a motion-free ground truth image. These experiments demonstrate a successful application of PVR motion compensation to the whole fetal body, uterus, and placenta.

Atlas-based approach for the segmentation of infant DTI MR brain images

In this paper, we propose a new adaptive atlas-based technique for the automated segmentation of brain tissues (white matter and grey matter) from infant diffusion tensor images (DTI). Brain images and desired region maps (brain, Cerebrospinal fluid, etc.) are modeled by a joint Markov-Gibbs random field (MGRF) model of independent image signals and interdependent region labels. The proposed joint MGRF model accounts for the following three descriptors: (i) a 1st-order visual appearance to describe the empirical distribution of six features that has been estimated from the DTI in addition to the non-diffusion (b0) scans, (ii) 3D probabilistic atlases, and (iii) a 3D spatially invariant 2nd-order homogeneity descriptor. The 1st-order visual appearance descriptor, assuming each of the estimated DTI parameters are independent, is precisely approximated using our previously developed linear combination of discrete Gaussians (LCDG) intensity model that includes positive and negative Gaussian components. The 3D probabilistic atlases are learned using a subset of the 3D co-aligned training DTI brain images. The 2nd-order homogeneity descriptor is modeled by a 2nd-order translation and rotation invariant MGRF of region labels, with analytically estimated potentials. We tested our approach on 25 DTI brain images, and evaluated the performance on 5 manually segmented 3D DTI brain images to confirm the high accuracy of the proposed approach, as evidenced by the Dice similarity, Hausdorff distance, and absolute volume difference metrics.