Groupwise Registration Research Articles

Retrospective motion correction for cardiac multi-parametric mapping with dictionary matching-based image synthesis and a low-rank constraint.

To develop a model-based motion correction (MoCo) method that does not need an analytical signal model to improve the quality of cardiac multi-parametric mapping. The proposed method constructs a hybrid loss that includes a dictionary-matching loss and a signal low-rankness loss, where the former registers the multi-contrast original images to a set of motion-free synthetic images and the latter forces the deformed images to be spatiotemporally coherent. We compared the proposed method with non-MoCo, a pairwise registration method (Pairwise-MI), and a groupwise registration method (pTVreg) via a free-breathing Multimapping dataset of 15 healthy subjects, both quantitatively and qualitatively. The proposed method achieved the lowest contour tracking errors (epicardium: 2.00 ± 0.39 mm vs 4.93 ± 2.29 mm, 3.50 ± 1.26 mm, and 2.61 ± 1.00 mm, and endocardium: 1.84 ± 0.34 mm vs 4.93 ± 2.40 mm, 3.43 ± 1.27 mm, and 2.55 ± 1.09 mm for the proposed method, non-MoCo, Pairwise-MI, and pTVreg, respectively; all p < 0.01) and the lowest dictionary matching errors among all methods. The proposed method also achieved the highest scores on the visual quality of mapping (T1: 4.74 ± 0.33 vs 2.91 ± 0.82, 3.58 ± 0.87, and 3.97 ± 1.05, and T2: 4.48 ± 0.56 vs 2.59 ± 0.81, 3.56 ± 0.93, and 4.14 ± 0.80 for the proposed method, non-MoCo, Pairwise-MI, and pTVreg, respectively; all p < 0.01). Finally, the proposed method had similar T1 and T2 mean values and SDs relative to the breath-hold reference in nearly all myocardial segments, whereas all other methods led to significantly different T1 and T2 measures and increases of SDs in multiple segments. The proposed method significantly improves the motion correction accuracy and mapping quality compared with non-MoCo and alternative image-based methods.

A Novel Registration Framework for Aligning Longitudinal Infant Brain Tensor Images.

Registering longitudinal infant brain images is challenging, as the infant brain undergoes rapid changes in size, shape and tissue contrast in the first months and years of life. Diffusion tensor images (DTI) have relatively consistent tissue properties over the course of infancy compared to commonly used T1 or T2-weighted images, presenting great potential for infant brain registration. Moreover, groupwise registration has been widely used in infant neuroimaging studies to reduce bias introduced by predefined atlases that may not be well representative of samples under study. To date, however, no methods have been developed for groupwise registration of tensor-based images. Here, we propose a novel registration approach to groupwise align longitudinal infant DTI images to a sample-specific common space. Longitudinal infant DTI images are first clustered into more homogenous subgroups based on image similarity using Louvain clustering. DTI scans are then aligned within each subgroup using standard tensor-based registration. The resulting images from all subgroups are then further aligned onto a sample-specific common space. Results show that our approach significantly improved registration accuracy both globally and locally compared to standard tensor-based registration and standard fractional anisotropy-based registration. Additionally, clustering based on image similarity yielded significantly higher registration accuracy compared to no clustering, but comparable registration accuracy compared to clustering based on chronological age. By registering images groupwise to reduce registration bias and capitalizing on the consistency of features in tensor maps across early infancy, our groupwise registration framework facilitates more accurate alignment of longitudinal infant brain images.

Coarse-to-fine registration and time-intensity curves constraint for liver DCE-MRI synthesis

Image registration plays a crucial role in dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI), used as a fundamental step for the subsequent diagnosis of benign and malignant tumors. However, the registration process encounters significant challenges due to the substantial intensity changes observed among different time points, resulting from the injection of contrast agents. Furthermore, previous studies have often overlooked the alignment of small structures, such as tumors and vessels. In this work, we propose a novel DCE-MRI registration framework that can effectively align the DCE-MRI time series. Specifically, our DCE-MRI registration framework consists of two steps, i.e., a de-enhancement synthesis step and a coarse-to-fine registration step. In the de-enhancement synthesis step, a disentanglement network separates DCE-MRI images into a content component representing the anatomical structures and a style component indicating the presence or absence of contrast agents. This step generates synthetic images where the contrast agents are removed from the original images, alleviating the negative effects of intensity changes on the subsequent registration process. In the registration step, we utilize a coarse registration network followed by a refined registration network. These two networks facilitate the estimation of both the coarse and refined displacement vector fields (DVFs) in a pairwise and groupwise registration manner, respectively. In addition, to enhance the alignment accuracy for small structures, a voxel-wise constraint is further conducted by assessing the smoothness of the time-intensity curves (TICs). Experimental results on liver DCE-MRI demonstrate that our proposed method outperforms state-of-the-art approaches, offering more robust and accurate alignment results.

ORRN: An ODE-Based Recursive Registration Network for Deformable Respiratory Motion Estimation With Lung 4DCT Images.

Deformable Image Registration (DIR) plays a significant role in quantifying deformation in medical data. Recent Deep Learning methods have shown promising accuracy and speedup for registering a pair of medical images. However, in 4D (3D + time) medical data, organ motion, such as respiratory motion and heart beating, can not be effectively modeled by pair-wise methods as they were optimized for image pairs but did not consider the organ motion patterns necessary when considering 4D data. This article presents ORRN, an Ordinary Differential Equations (ODE)-based recursive image registration network. Our network learns to estimate time-varying voxel velocities for an ODE that models deformation in 4D image data. It adopts a recursive registration strategy to progressively estimate a deformation field through ODE integration of voxel velocities. We evaluate the proposed method on two publicly available lung 4DCT datasets, DIRLab and CREATIS, for two tasks: 1) registering all images to the extreme inhale image for 3D+t deformation tracking and 2) registering extreme exhale to inhale phase images. Our method outperforms other learning-based methods in both tasks, producing the smallest Target Registration Error of 1.24 mm and 1.26 mm, respectively. Additionally, it produces less than 0.001% unrealistic image folding, and the computation speed is less than 1 s for each CT volume. ORRN demonstrates promising registration accuracy, deformation plausibility, and computation efficiency on group-wise and pair-wise registration tasks. It has significant implications in enabling fast and accurate respiratory motion estimation for treatment planning in radiation therapy or robot motion planning in thoracic needle insertion.

Improved fMRI-based pain prediction using Bayesian group-wise functional registration.

In recent years, the field of neuroimaging has undergone a paradigm shift, moving away from the traditional brain mapping approach towards the development of integrated, multivariate brain models that can predict categories of mental events. However, large interindividual differences in both brain anatomy and functional localization after standard anatomical alignment remain a major limitation in performing this type of analysis, as it leads to feature misalignment across subjects in subsequent predictive models. This article addresses this problem by developing and validating a new computational technique for reducing misalignment across individuals in functional brain systems by spatially transforming each subject's functional data to a common latent template map. Our proposed Bayesian functional group-wise registration approach allows us to assess differences in brain function across subjects and individual differences in activation topology. We achieve the probabilistic registration with inverse-consistency by utilizing the generalized Bayes framework with a loss function for the symmetric group-wise registration. It models the latent template with a Gaussian process, which helps capture spatial features in the template, producing a more precise estimation. We evaluate the method in simulation studies and apply it to data from an fMRI study of thermal pain, with the goal of using functional brain activity to predict physical pain. We find that the proposed approach allows for improved prediction of reported pain scores over conventional approaches. Received on 2 January 2017. Editorial decision on 8 June 2021.

Sub-voxel quantitative susceptibility mapping for assessing whole-brain magnetic susceptibility from ages 4 to 80.

The evolution of magnetic susceptibility of the brain is mainly determined by myelin in white matter (WM) and iron deposition in deep gray matter (DGM). However, existing imaging techniques have limited abilities to simultaneously quantify the myelination and iron deposition within a voxel throughout brain development and aging. For instance, the temporal trajectories of iron in the brain WM and myelination in DGM have not been investigated during the aging process. This study aimed to map the age-related iron and myelin changes in the whole brain, encompassing myelin in DGM and iron deposition in WM, using a novel sub-voxel quantitative susceptibility mapping (QSM) method. To achieve this, a cohort of 494 healthy adults (18-80 years old) was studied. The sub-voxel QSM method was employed to obtain the paramagnetic and diamagnetic susceptibility based on the approximated map from acquired map. The linear relationship between and maps was established from the regression coefficients on a small cohort data acquired with both 3D gradient recalled echo data and mapping. Large cohort sub-voxel susceptibility maps were used to create longitudinal and age-specific atlases via group-wise registration. To explore the differential developmental trajectories in the DGM and WM, we employed nonlinear models including exponential and Poisson functions, along with generalized additive models. The constructed atlases reveal the iron accumulation in the posterior part of the putamen and the gradual myelination process in the globus pallidus with aging. Interestingly, the developmental trajectories show that the rate of myelination differs among various DGM regions. Furthermore, the process of myelin synthesis is paralleled by an associated pattern of iron accumulation in the primary WM fiber bundles. In summary, our study offers significant insights into the distinctive developmental trajectories of iron in the brain's WM and myelination/demyelination in the DGM in vivo. These findings highlight the potential of using sub-voxel QSM to uncover new perspectives in neuroscience and improve our understanding of whole-brain myelination and iron deposit processes across the lifespan.

X-Metric: An N-Dimensional Information-Theoretic Framework for Groupwise Registration and Deep Combined Computing.

This paper presents a generic probabilistic framework for estimating the statistical dependency and finding the anatomical correspondences among an arbitrary number of medical images. The method builds on a novel formulation of the N-dimensional joint intensity distribution by representing the common anatomy as latent variables and estimating the appearance model with nonparametric estimators. Through connection to maximum likelihood and the expectation-maximization algorithm, an information-theoretic metric called X-metric and a co-registration algorithm named X-CoReg are induced, allowing groupwise registration of the N observed images with computational complexity of O(N). Moreover, the method naturally extends for a weakly-supervised scenario where anatomical labels of certain images are provided. This leads to a combined-computing framework implemented with deep learning, which performs registration and segmentation simultaneously and collaboratively in an end-to-end fashion. Extensive experiments were conducted to demonstrate the versatility and applicability of our model, including multimodal groupwise registration, motion correction for dynamic contrast enhanced magnetic resonance images, and deep combined computing for multimodal medical images. Results show the superiority of our method in various applications in terms of both accuracy and efficiency, highlighting the advantage of the proposed representation of the imaging process.

Convolution-based free-form deformation for multimodal groupwise registration

Recently, an efficient implementation of convolution-based free form deformations (FFD) has been proposed for both groupwise 3D monomodal and 2D pairwise multimodal registrations. However, there is still an unmet need in the field for groupwise L-D multimodal registration with L≥2. In this correspondence, we address this need and present a solution for achieving accurate registration using two popular metrics: Renyi entropy and PCA2.

Free-breathing simultaneous native myocardial T1, T2 and T1ρ mapping with Cartesian acquisition and dictionary matching

BackgroundT1, T2 and T1ρ are well-recognized parameters for quantitative cardiac MRI. Simultaneous estimation of these parameters allows for comprehensive myocardial tissue characterization, such as myocardial fibrosis and edema. However, conventional techniques either quantify the parameters individually with separate breath-hold acquisitions, which may result in unregistered parameter maps, or estimate multiple parameters in a prolonged breath-hold acquisition, which may be intolerable to patients. We propose a free-breathing multi-parametric mapping (FB-MultiMap) technique that provides co-registered myocardial T1, T2 and T1ρ maps in a single efficient acquisition. MethodsThe proposed FB-MultiMap performs electrocardiogram-triggered single-shot Cartesian acquisition over 16 consecutive cardiac cycles, where inversion, T2 and T1ρ preparations are introduced for varying contrasts. A diaphragmatic navigator was used for prospective through-plane motion correction and the in-plane motion was corrected retrospectively with a group-wise image registration method. Quantitative mapping was conducted through dictionary matching of the motion corrected images, where the subject-specific dictionary was created using Bloch simulations for a range of T1, T2 and T1ρ values, as well as B1 factors to account for B1 inhomogeneities. The FB-MultiMap was optimized and validated in numerical simulations, phantom experiments, and in vivo imaging of 15 healthy subjects and six patients with suspected cardiac diseases. ResultsThe phantom T1, T2 and T1ρ values estimated with FB-MultiMap agreed well with reference measurements with no dependency on heart rate. In healthy subjects, FB-MultiMap T1 was higher than MOLLI T1 mapping (1218 ± 50 ms vs. 1166 ± 38 ms, p < 0.001). The myocardial T2 and T1ρ estimated with FB-MultiMap were lower compared to the mapping with T2- or T1ρ-prepared 2D balanced steady-state free precession (T2: 41.2 ± 2.8 ms vs. 42.5 ± 3.1 ms, p = 0.06; T1ρ: 45.3 ± 4.4 ms vs. 50.2 ± 4.0, p < 0.001). The pathological changes in myocardial parameters measured with FB-MultiMap were consistent with conventional techniques in all patients. ConclusionThe proposed free-breathing multi-parametric mapping technique provides co-registered myocardial T1, T2 and T1ρ maps in 16 heartbeats, achieving similar mapping quality to conventional breath-hold mapping methods.

Fuzzy Cluster-based Group-wise Point Set Registration with Quality Assessment.

This article studies group-wise point set registration and makes the following contributions: "FuzzyGReg", which is a new fuzzy cluster-based method to register multiple point sets jointly, and "FuzzyQA", which is the associated quality assessment to check registration accuracy automatically. Given a group of point sets, FuzzyGReg creates a model of fuzzy clusters and equally treats all the point sets as the elements of the fuzzy clusters. Then, the group-wise registration is turned into a fuzzy clustering problem. To resolve this problem, FuzzyGReg applies a fuzzy clustering algorithm to identify the parameters of the fuzzy clusters while jointly transforming all the point sets to achieve an alignment. Next, based on the identified fuzzy clusters, FuzzyQA calculates the spatial properties of the transformed point sets and then checks the alignment accuracy by comparing the similarity degrees of the spatial properties of the point sets. When a local misalignment is detected, a local re-alignment is performed to improve accuracy. The proposed method is cost-efficient and convenient to be implemented. In addition, it provides reliable quality assessments in the absence of ground truth and user intervention. In the experiments, different point sets are used to test the proposed method and make comparisons with state-of-the-art registration techniques. The experimental results demonstrate the effectiveness of our method. The code is available at https://gitsvn-nt.oru.se/qianfang.liao/FuzzyGRegWithQA.

GMCNet: A Generative Multi-Resolution Framework for Cardiac Registration

Deformable image registration plays a crucial role in estimating cardiac deformation from a sequence of images. However, existing registration methods primarily process images as pairs instead of processing all images in a sequence together. This study proposes a novel end-to-end learning-free generative multi-resolution convolutional neural network (GMCNet) with the primary focus of registering images in a sequence. Even though learning-based methods have yielded high performance for image registration, their performance depends on their ability to learn information from a large number of samples which are difficult to obtain and might bias the framework to the specific domain of data. The proposed learning-free method eliminates the need for a dedicated training set while exploiting the capabilities of neural networks to achieve accurate deformation fields. Due to its capability of parameter sharing through the architecture, the GMCNet can be used as a groupwise registration as well as pairwise registration. The proposed method was evaluated on three different clinical cardiac magnetic resonance imaging datasets and compared quantitatively against nine other state-of-the-art learning and optimization-based algorithms. The proposed method outperformed other methods in all comparisons and yielded average Dice metric values ranging from 0.85 to 0.88 for the datasets. Different aspects of the GMCNet are also explored by assessing 1) the robustness; 2) performance on pairwise registration; 3) the influence of spatial transformation in a controlled environment; and 4) the impact of different multi-resolution structures. The results demonstrate that using temporal information to estimate the deformation fields leads to more accurate registration results and improved robustness under different noise levels. Moreover, the proposed method does not need images for training, and therefore, its prediction is not domain-specific and can be applied to any sequence of images.

Spatiotemporal Atlas of the Fetal Brain Depicts Cortical Developmental Gradient.

The fetal brains experience rapid and complex development in utero during the second and third trimesters. In utero MRI of the fetal brain in this period enables us to quantify normal fetal brain development in the spatiotemporal domain. In this study, we established a high-quality spatiotemporal atlas between 23 and 38 weeks gestational age (GA) from 90 healthy Chinese human fetuses of both sexes using a pairwise and groupwise registration pipeline. We quantified the fetal cortical morphology indices and characterized their spatiotemporal developmental pattern. The cortical thickness exhibited a biphasic pattern that first increased and then decreased; the curvature fitted well into the Gompertz growth model; sulcal depth increased linearly, while surface area expanded exponentially. The cortical thickness and curvature trajectories consistently pointed to a characteristic time point around GA of 31 weeks. The characteristic GA and growth rate obtained from individual cortical regions suggested a central-to-peripheral developmental gradient, with the earliest development in the parietal lobe, and we also observed a superior-to-inferior gradient within the temporal lobe. These findings may be linked to biophysical events, such as dendritic arborization and thalamocortical fibers ingrowth. The proposed atlas was also compared with an existing fetal atlas from a white/mixed population. Finally, we examined the structural asymmetry of the fetal brains and found extensive asymmetry that dynamically changed with development. The current study depicted a comprehensive profile of fetal cortical development, and the established atlas could be used as a normative reference for neurodevelopmental and diagnostic purposes, especially in the Chinese population.SIGNIFICANCE STATEMENT We generated a high-quality 4D spatiotemporal atlas of the normal fetal brain development from 23 to 38 gestational weeks in a Chinese population and characterized the spatiotemporal developmental pattern of cortical morphology. According to the cortical development trajectories, the fetal cerebral cortex development follows a central-to-peripheral developmental gradient that may be related to the underlying cellular events. The majority of cortical regions already exhibit significant asymmetry during the fetal period.

Quantitative susceptibility atlas construction in Montreal Neurological Institute space: towards histological-consistent iron-rich deep brain nucleus subregion identification.

Iron-rich deep brain nuclei (DBN) of the human brain are involved in various motoric, emotional and cognitive brain functions. The abnormal iron alterations in the DBN are closely associated with multiple neurological and psychiatric diseases. Quantitative susceptibility mapping (QSM) provides the spatial distribution of the magnetic susceptibility of human brain tissues. Compared to traditional structural imaging, QSM provides superiority for imaging the iron-rich DBN owing to the susceptibility difference existing between brain tissues. In this study, we constructed a Montreal Neurological Institute (MNI) space unbiased QSM human brain atlas via group-wise registration from 100 healthy subjects aged 19-29 years. The atlas construction process was guided by hybrid images that were fused from multi-modal magnetic resonance images (MRI). We named it as Multi-modal-fused magnetic Susceptibility (MuSus-100) atlas. The high-quality susceptibility atlas provides extraordinary image contrast between iron-rich DBN with their surroundings. Parcellation maps of DBN and their subregions that are highly related to neurological and psychiatric pathology were then manually labeled based on the atlas set with the assistance of an image border-enhancement process. Especially, the bilateral thalamus was delineated into 64 detailed subregions referring to the Schaltenbrand-Wahren stereotactic atlas. To our best knowledge, the histological-consistent thalamic nucleus parcellation map is well defined for the first time in the MNI space. Compared with existing atlases that emphasizing DBN parcellation, the newly proposed atlas outperforms on the task of atlas-guided individual brain image DBN segmentation both in accuracy and robustness. Moreover, we applied the proposed DBN parcellation map to conduct detailed identification of the pathology-related iron content alterations in subcortical nuclei for Parkinson's Disease (PD) patients. We envision that the MuSus-100 atlas can play a crucial role in improving the accuracy of DBN segmentation for the research of neurological and psychiatric disease progress and also be helpful for target planning in deep brain stimulation surgery.

Automated mitral annulus annotation over 4D-CT for transcatheter mitral valve replacement

ABSTRACT The morphological assessment of the geometry and the deformation of the mitral annulus has the potential to support the development of minimally invasive solutions for patients suffering from mitral valve regurgitation. Cardiac 4D-CT allows such a morphological assessment, which is, however, often performed manually, by annotating the mitral annulus and other cardiac landmarks. The goal of this study is to automate the annotation of cardiac landmarks in 4D-CT by propagating them from one reference phase to all others. The method proposed is based on non-rigid image registration, and four alternative variations have been tested, based on pairwise and groupwise registration. Their accuracy is evaluated against inter-observer variability on eight 4D-CTs. The results show that the pairwise and groupwise approaches using independent 3D transforms have a comparable performance, with an accuracy comparable with the inter-observer variability; however, the latter approach was computationally more expensive. Instead, the spatio-temporal groupwise approach obtained significantly worse results. The technology has, therefore, the potential to make an efficient analysis of the deforming heart more attainable. Moreover, the manual annotation process confirms that the mitral annulus deforms significantly during the cardiac contraction, supporting the use of 4D imaging for a reliable morphological assessment.

A groupwise registration and tractography framework for cardiac myofiber architecture description by diffusion MRI: An application to the ventricular junctions.

BackgroundKnowledge of the normal myocardial–myocyte orientation could theoretically allow the definition of relevant quantitative biomarkers in clinical routine to diagnose heart pathologies. A whole heart diffusion tensor template representative of the global myofiber organization over species is therefore crucial for comparisons across populations. In this study, we developed a groupwise registration and tractography framework to resolve the global myofiber arrangement of large mammalian sheep hearts. To demonstrate the potential application of the proposed method, a novel description of sub-regions in the intraventricular septum is presented.MethodsThree explanted sheep (ovine) hearts (size ~12×8×6 cm3, heart weight ~ 150 g) were perfused with contrast agent and fixative and imaged in a 9.4T magnet. A group-wise registration of high-resolution anatomical and diffusion-weighted images were performed to generate anatomical and diffusion tensor templates. Diffusion tensor metrics (eigenvalues, eigenvectors, fractional anisotropy …) were computed to provide a quantitative and spatially-resolved analysis of cardiac microstructure. Then tractography was performed using deterministic and probabilistic algorithms and used for different purposes: i) Visualization of myofiber architecture, ii) Segmentation of sub-area depicting the same fiber organization, iii) Seeding and Tract Editing. Finally, dissection was performed to confirm the existence of macroscopic structures identified in the diffusion tensor template.ResultsThe template creation takes advantage of high-resolution anatomical and diffusion-weighted images obtained at an isotropic resolution of 150 μm and 600 μm respectively, covering ventricles and atria and providing information on the normal myocardial architecture. The diffusion metric distributions from the template were found close to the one of the individual samples validating the registration procedure. Small new sub-regions exhibiting spatially sharp variations in fiber orientation close to the junctions of the septum and ventricles were identified. Each substructure was defined and represented using streamlines. The existence of a fiber-bundles in the posterior junction was validated by anatomical dissection. A complex structural organization of the anterior junction in comparison to the posterior junction was evidenced by the high-resolution acquisition.ConclusionsA new framework combining cardiac template generation and tractography was applied on the whole sheep heart. The framework can be used for anatomical investigation, characterization of microstructure and visualization of myofiber orientation across samples. Finally, a novel description of the ventricular junction in large mammalian sheep hearts was proposed.

Joint Deformable Image Registration and ADC Map Regularization: Application to DWI-Based Lymphoma Classification.

The Apparent Diffusion Coefficient (ADC) is considered an importantimaging biomarker contributing to the assessment of tissue microstructure and pathophy- siology. It is calculated from Diffusion-Weighted Magnetic Resonance Imaging (DWI) by means of a diffusion model, usually without considering any motion during image acquisition. We propose a method to improve the computation of the ADC by coping jointly with both motion artifacts in whole-body DWI (through group-wise registration) and possible instrumental noise in the diffusion model. The proposed deformable registration method yielded on average the lowest ADC reconstruction error on data with simulated motion and diffusion. Moreover, our approach was applied on whole-body diffusion weighted images obtained with five different b-values from a cohort of 38 patients with histologically confirmed lymphomas of three different types (Hodgkin, diffuse large B-cell lymphoma and follicular lymphoma). Evaluation on the real data showed that ADC-based features, extracted using our joint optimization approach classified lymphomas with an accuracy of approximately 78.6% (yielding a 11% increase in respect to the standard features extracted from unregistered diffusion-weighted images). Furthermore, the correlation between diffusion characteristics and histopathological findings was higher than any other previous approach of ADC computation.

Saturated multi‐delay renal arterial spin labeling technique for simultaneous perfusion and T1 quantification in kidneys

To propose a free-breathing simultaneous multi-delay arterial spin labeling (ASL) and T1 mapping technique with a stepwise kinetic model for renal assessment in a single 4-min scan at 3T. The proposed saturated multi-delay renal arterial spin labeling (SAMURAI) sequence used flow-sensitive alternating inversion recovery (FAIR) preparation, followed by acquisition of 9 images with Look-Locker spoiled gradient recalled echo (SPGR). Pre-saturation at the imaging slice was used to achieve saturation-based T1 mapping. A 4-step 2-compartment kinetic model was proposed to characterize water transition through artery- and tissue-compartment. The impact of the Look-Locker sampling scheme on the ASL signal was corrected in this model. T1 estimation with dictionary searching method and perfusion quantification based on the proposed kinetic model fitting were conducted after groupwise registration of the acquired images. The feasibility and repeatability of SAMURAI were validated in healthy subjects (n=11) and patients with different renal diseases (n=4). The proposed SAMURAI technique can provide accurate T1 map with strong correlation (R2 =0.98) with inversion recovery spin echo (IR-SE) on phantom. SAMURAI provided equally reliable whole kidney and cortical ASL and T1 quantification results compared with multi-TI FAIR (intraclass correlation coefficient [ICC], 0.880-0.958) and IR-SPGR (ICC, 0.875-0.912), respectively. Low renal blood flow and increased T1 were detected by SAMURAI in the affected kidneys of the patients. SAMURAI had excellent scan-rescan repeatability (ICC, 0.905-0.992) and significantly reduced scan time (4 min 6 s vs. 45 min for 9 TIs) compared to multi-TI FAIR. The proposed SAMURAI technique is feasible and repeatable for simultaneously quantifying T1 and perfusion of kidneys with high time-efficiency.

Multi-View 3D Transesophageal Echocardiography Registration and Volume Compounding for Mitral Valve Procedure Planning

Three-dimensional ultrasound mosaicing can increase image quality and expand the field of view. However, limited work has been done applying these compounded approaches for cardiac procedures focused on the mitral valve. For procedures targeting the mitral valve, transesophageal echocardiography (TEE) is the primary imaging modality used as it provides clear 3D images of the valve and surrounding tissues. However, TEE suffers from image artefacts and signal dropout, particularly for structures lying below the valve, including chordae tendineae, making it necessary to acquire alternative echo views to visualize these structures. Due to the limited field of view obtainable, the entire ventricle cannot be directly visualized in sufficient detail from a single image acquisition in 3D. We propose applying an image compounding technique to TEE volumes acquired from a mid-esophageal position and several transgastric positions in order to reconstruct a high-detail volume of the mitral valve and sub-valvular structures. This compounding technique utilizes both fully and semi-simultaneous group-wise registration to align the multiple 3D volumes, followed by a weighted intensity compounding step based on the monogenic signal. This compounding technique is validated using images acquired from two excised porcine mitral valve units and three patient data sets. We demonstrate that this compounding technique accurately captures the physical structures present, including the mitral valve, chordae tendineae and papillary muscles. The chordae length measurement error between the compounded ultrasound and ground-truth CT for two porcine valves is reported as 0.7 ± 0.6 mm and 0.6 ± 0.6 mm.

End-to-End Deep Learning of Non-rigid Groupwise Registration and Reconstruction of Dynamic MRI.

Temporal correlation has been exploited for accelerated dynamic MRI reconstruction. Some methods have modeled inter-frame motion into the reconstruction process to produce temporally aligned image series and higher reconstruction quality. However, traditional motion-compensated approaches requiring iterative optimization of registration and reconstruction are time-consuming, while most deep learning-based methods neglect motion in the reconstruction process. We propose an unrolled deep learning framework with each iteration consisting of a groupwise diffeomorphic registration network (GRN) and a motion-augmented reconstruction network. Specifically, the whole dynamic sequence is registered at once to an implicit template which is used to generate a new set of dynamic images to efficiently exploit the full temporal information of the acquired data via the GRN. The generated dynamic sequence is then incorporated into the reconstruction network to augment the reconstruction performance. The registration and reconstruction networks are optimized in an end-to-end fashion for simultaneous motion estimation and reconstruction of dynamic images. The effectiveness of the proposed method is validated in highly accelerated cardiac cine MRI by comparing with other state-of-the-art approaches.

Lung-CRNet: A convolutional recurrent neural network for lung 4DCT image registration.

Deformable image registration (DIR) of lung four-dimensional computed tomography (4DCT) plays a vital role in a wide range of clinical applications. Most of the existing deep learning-based lung 4DCT DIR methods focus on pairwise registration which aims to register two images with large deformation. However, the temporal continuities of deformation fields between phases are ignored. This paper proposes a fast and accurate deep learning-based lung 4DCT DIR approach that leverages the temporal component of 4DCT images. We present Lung-CRNet, an end-to-end convolutional recurrent registration neural network for lung 4DCT images and reformulate 4DCT DIR as a spatiotemporal sequence predicting problem in which the input is a sequence of three-dimensional computed tomography images from the inspiratory phase to the expiratory phase in a respiratory cycle. The first phase in the sequence is selected as the only reference image and the rest as moving images. Multiple convolutional gated recurrent units (ConvGRUs) are stacked to capture the temporal clues between images. The proposed network is trained in an unsupervised way using a spatial transformer layer. During inference, Lung-CRNet is able to yield the respective displacement field for each reference-moving image pair in the input sequence. We have trained the proposed network using a publicly available lung 4DCT dataset and evaluated performance on the widely used the DIR-Lab dataset. The mean and standard deviation of target registration error are 1.56 1.05 mm on the DIR-Lab dataset. The computation time for each forward prediction is less than 1 s on average. The proposed Lung-CRNet is comparable to the existing state-of-the-art deep learning-based 4DCT DIR methods in both accuracy and speed. Additionally, the architecture of Lung-CRNet can be generalized to suit other groupwise registration tasks which align multiple imagessimultaneously.