This paper presents an efficient algorithm for large deformation diffeomorphic metric mapping (LDDMM) with geodesic shooting for image registration. We introduce a novel finite dimensional Fourier representation of diffeomorphic deformations based on the key fact that the high frequency components of a diffeomorphism remain stationary throughout the integration process when computing the deformation associated with smooth velocity fields. We show that manipulating high dimensional diffeomorphisms can be carried out entirely in the bandlimited space by integrating the nonstationary low frequency components of the displacement field. This insight substantially reduces the computational cost of the registration problem. Experimental results show that our method is significantly faster than the state-of-the-art diffeomorphic image registration methods while producing equally accurate alignment. We demonstrate our algorithm in two different applications of image registration: neuroimaging and in-utero imaging.
We present an algorithm for creating high resolution anatomically plausible images consistent with acquired clinical brain MRI scans with large inter-slice spacing. Although large databases of clinical images contain a wealth of information, medical acquisition constraints result in sparse scans that miss much of the anatomy. These characteristics often render computational analysis impractical as standard processing algorithms tend to fail when applied to such images. Highly specialized or application-specific algorithms that explicitly handle sparse slice spacing do not generalize well across problem domains. In contrast, our goal is to enable application of existing algorithms that were originally developed for high resolution research scans to significantly undersampled scans. We introduce a model that captures fine-scale anatomical similarity across subjects in clinical image collections and use it to fill in the missing data in scans with large slice spacing. Our experimental results demonstrate that the proposed method outperforms current upsampling methods and promises to facilitate subsequent analysis not previously possible with scans of this quality.
Diffusion tensor imaging (DTI) studies in chronic schizophrenia have found widespread but often inconsistent patterns of white matter abnormalities. These studies have typically used the conventional measure of fractional anisotropy, which can be contaminated by extracellular free-water. A recent free-water imaging study reported reduced free-water corrected fractional anisotropy (FAT) in chronic schizophrenia across several brain regions, but limited changes in the extracellular volume. The present study set out to validate these findings in a substantially larger sample. Tract-based spatial statistics (TBSS) was performed in 188 healthy controls and 281 chronic schizophrenia patients. Forty-two regions of interest (ROIs), as well as average whole-brain FAT and FW were extracted from free-water corrected diffusion tensor maps. Compared to healthy controls, reduced FAT was found in the chronic schizophrenia group in the anterior limb of the internal capsule bilaterally, the posterior thalamic radiation bilaterally, as well as the genu and body of the corpus callosum. While a significant main effect of group was observed for FW, none of the follow-up contrasts survived correction for multiple comparisons. The observed FAT reductions in the absence of extracellular FW changes, in a large, multi-site sample of chronic schizophrenia patients, validate the pattern of findings reported by a previous, smaller free-water imaging study of a similar sample. The limited number of regions in which FAT was reduced in the schizophrenia group suggests that actual white matter tissue degeneration in chronic schizophrenia, independent of extracellular FW, might be more localized than suggested previously.
We propose an extension of the Wasserstein 1-metric (W1) for density matrices, matrix-valued density measures, and an unbalanced interpretation of mass transport. We use duality theory and, in particular, a "dual of the dual" formulation of W1. This matrix analogue of the Earth Mover's Distance has several attractive features including ease of computation.
Tractography based on non-invasive diffusion imaging is central to the study of human brain connectivity. To date, the approach has not been systematically validated in ground truth studies. Based on a simulated human brain data set with ground truth tracts, we organized an open international tractography challenge, which resulted in 96 distinct submissions from 20 research groups. Here, we report the encouraging finding that most state-of-the-art algorithms produce tractograms containing 90% of the ground truth bundles (to at least some extent). However, the same tractograms contain many more invalid than valid bundles, and half of these invalid bundles occur systematically across research groups. Taken together, our results demonstrate and confirm fundamental ambiguities inherent in tract reconstruction based on orientation information alone, which need to be considered when interpreting tractography and connectivity results. Our approach provides a novel framework for estimating reliability of tractography and encourages innovation to address its current limitations.
OBJECTIVE: We introduce descriptor-based segmentation that extends existing patch-based methods by combining intensities, features, and location information. Since it is unclear which image features are best suited for patch selection, we perform a broad empirical study on a multitude of different features. METHODS: We extend nonlocal means segmentation by including image features and location information. We search larger windows with an efficient nearest neighbor search based on kd-trees. We compare a large number of image features. RESULTS: The best results were obtained for entropy image features, which have not yet been used for patch-based segmentation. We further show that searching larger image regions with an approximate nearest neighbor search and location information yields a significant improvement over the bounded nearest neighbor search traditionally employed in patch-based segmentation methods. CONCLUSION: Features and location information significantly increase the segmentation accuracy. The best features highlight boundaries in the image. SIGNIFICANCE: Our detailed analysis of several aspects of nonlocal means-based segmentation yields new insights about patch and neighborhood sizes together with the inclusion of location information. The presented approach advances the state-of-the-art in the segmentation of parotid glands for radiation therapy planning.
Quantitative analysis of clinical image data is an active area of research that holds promise for precision medicine, early assessment of treatment response, and objective characterization of the disease. Interoperability, data sharing, and the ability to mine the resulting data are of increasing importance, given the explosive growth in the number of quantitative analysis methods being proposed. The Digital Imaging and Communications in Medicine (DICOM) standard is widely adopted for image and metadata in radiology. dcmqi (DICOM for Quantitative Imaging) is a free, open source library that implements conversion of the data stored in commonly used research formats into the standard DICOM representation. dcmqi source code is distributed under BSD-style license. It is freely available as a precompiled binary package for every major operating system, as a Docker image, and as an extension to 3D Slicer. Installation and usage instructions are provided on Harvard DASH.
The Neuroimage Analysis Center's Computational Clinical Anatomy Core and the Surgical Planning Laboratory at Brigham and Women's Hospital is pleased to make available a multi-modality MRI-based atlas of the brain. Data was acquired at the Martinos Center for Biomedical Imaging (courtesy Dr. Lawrence Wald) on a Siemens 3T scanner, using a multi-array head coil, in a healthy, 42 year old male. The data set consists of : 1. a volumetric whole head MPRAGE series (voxel size 0.75 mm isotropic). 2. a volumetric whole head T2-weighted series (voxel size 0.75 mm isotropic). 3. a downsampled version of both acquisitions at 1mm isotropic resolution. 4. a per voxel labeling of the structures based on the 1mm volumes. 5. a color file mapping label values to RadLex-ontology derived names and colors suitable for display. 6. MRML files for displaying the volumes in 3D Slicer version 3.6 or greater, available for download. The atlas data is made available under terms of the 3D Slicer License section B. The Slicer4 version also consists of 1. hypotalamic parcellation (courtesy Nikos Makris [Neuroimage. 2013]) 2. cerebellar parcellation (courtesy Nikos Makris [J Cogn Neurosci. 2003], [Neuroimage. 2005]) 3.head and neck muscles segmentation 4. anatomical model hierarchy 5. several pre-defined Scene Views (“anatomy teaching files”). All in a mrb (Medical Reality Bundle) archive file that contains the mrml scene file and all data for loading into Slicer 4 for displaying the volumes in 3D Slicer version 4.0 or greater, available for download. This work is funded as part of the Neuroimaging Analysis Center, grant number P41 RR013218, by the NIH's National Center for Research Resources (NCRR) and grant number P41 EB015902, by the NIH's National Institute of Biomedical Imaging and Bioengineering (NIBIB) and the Google Faculty Research Award. Contributors: Ilwoo Lyu and Martin Styner: Sulcal Curves, Samira Farough: Ventricular System, Ibraheem Naeem and Maria Naeem: Head and Neck Muscles, George Papadimitriou: Cerebellar Parcellation, Madiha Tahir: White Matter. This atlas maybe viewed with our Open Anatomy Browser.
Segmentation is a fundamental task for extracting semantically meaningful regions from an image. The goal of segmentation algorithms is to accurately assign object labels to each image location. However, image noise, shortcomings of algorithms, and image ambiguities cause uncertainty in label assignment. Estimating this uncertainty is important in multiple application domains, such as segmenting tumors from medical images for radiation treatment planning. One way to estimate these uncertainties is through the computation of posteriors of Bayesian models, which is computationally prohibitive for many practical applications. However, most computationally efficient methods fail to estimate label uncertainty. We therefore propose in this paper the active mean fields (AMF) approach, a technique based on Bayesian modeling that uses a mean-field approximation to efficiently compute a segmentation and its corresponding uncertainty. Based on a variational formulation, the resulting convex model combines any label-likelihood measure with a prior on the length of the segmentation boundary. A specific implementation of that model is the Chan-Vese segmentation model, in which the binary segmentation task is defined by a Gaussian likelihood and a prior regularizing the length of the segmentation boundary. Furthermore, the Euler-Lagrange equations derived from the AMF model are equivalent to those of the popular Rudin-Osher-Fatemi (ROF) model for image denoising. Solutions to the AMF model can thus be implemented by directly utilizing highly efficient ROF solvers on log-likelihood ratio fields. We qualitatively assess the approach on synthetic data as well as on real natural and medical images. For a quantitative evaluation, we apply our approach to the tt icgbench dataset.
We consider transportation over a strongly connected, directed graph. The scheduling amounts to selecting transition probabilities for a discrete-time Markov evolution which is designed to be consistent with initial and final marginal constraints on mass transport. We address the situation where initially the mass is concentrated on certain nodes and needs to be transported in a certain time period to another set of nodes, possibly disjoint from the first. The random evolution is selected to be closest to a prior measure on paths in the relative entropy sense-such a construction is known as a Schrödinger bridge between the two given marginals. It may be viewed as an atypical stochastic control problem where the control consists in suitably modifying the prior transition mechanism. The prior can be chosen to incorporate constraints and costs for traversing specific edges of the graph, but it can also be selected to allocate equal probability to all paths of equal length connecting any two nodes (i.e., a uniform distribution on paths). This latter choice for prior transitions relies on the so-called Ruelle-Bowen random walker and gives rise to scheduling that tends to utilize all paths as uniformly as the topology allows. Thus, this Ruelle-Bowen law () taken as prior, leads to a transportation plan that tends to lessen congestion and ensures a level of robustness. We also show that the distribution on paths, which attains the maximum entropy rate for the random walker given by the topological entropy, can itself be obtained as the time-homogeneous solution of a maximum entropy problem for measures on paths (also a Schrödinger bridge problem, albeit with prior that is not a probability measure). Finally we show that the paradigm of Schrödinger bridges as a mechanism for scheduling transport on networks can be adapted to graphs that are not strongly connected, as well as to weighted graphs. In the latter case, our approach may be used to design a transportation plan which effectively compromises between robustness and other criteria such as cost. Indeed, we explicitly provide a robust transportation plan which assigns maximum probability to minimum cost paths and therefore compares favourably with Optimal Mass Transportation strategies.
We propose an optimization method for estimating patient- specific muscle fiber arrangement from clinical CT. Our approach first computes the structure tensor field to estimate local orientation, then a geometric template representing fiber arrangement is fitted using a B- spline deformation by maximizing fitness of the local orientation using a smoothness penalty. The initialization is computed with a previously proposed algorithm that takes account of only the muscle’s surface shape. Evaluation was performed using a CT volume (1.0mm3/voxel) and high resolution optical images of a serial cryosection (0.1mm3/voxel). The mean fiber distance error at the initialization of 6.00 mm was decreased to 2.78mm after the proposed optimization for the gluteus maximus muscle, and from 5.28 mm to 3.09 mm for the gluteus medius muscle. The result from 20 patient CT images suggested that the proposed algorithm reconstructed an anatomically more plausible fiber arrangement than the previous method.
Geodesic regression on images enables studies of brain development and degeneration, disease progression, and tumor growth. The high-dimensional nature of image data presents significant computational challenges for the current regression approaches and prohibits large scale studies. In this paper, we present a fast geodesic regression method that dramatically decreases the computational cost of the inference procedure while maintaining prediction accuracy. We employ an efficient low dimensional representation of diffeomorphic transformations derived from the image data and characterize the regressed trajectory in the space of diffeomorphisms by its initial conditions, i.e., an initial image template and an initial velocity field computed as a weighted average of pairwise diffeomorphic image registration results. This construction is achieved by using a first-order approximation of pairwise distances between images. We demonstrate the efficiency of our model on a set of 3D brain MRI scans from the OASIS dataset and show that it is dramatically faster than the state-of-the-art regression methods while producing equally good regression results on the large subject cohort.
We propose a general dynamic regression framework for partial correlation and causality analysis of functional brain networks. Using the optimal prediction theory, we present the solution of the dynamic regression problem by minimizing the entropy of the associated stochastic process. We also provide the relation between the solutions and the linear dependence models of Geweke and Granger and derive novel expressions for computing partial correlation and causality using an optimal prediction filter with minimum error variance. We use the proposed dynamic framework to study the intrinsic partial correlation and causal- ity between seven different brain networks using resting state functional MRI (rsfMRI) data from the Human Connectome Project (HCP) and compare our results with those obtained from standard correlation and causality measures. The results show that our optimal prediction filter explains a significant portion of the variance in the rsfMRI data at low frequencies, unlike standard partial correlation analysis.
This work presents a supra-threshold fiber cluster (STFC) analysis that leverages the whole brain fiber geometry to enhance sta- tistical group difference analysis. The proposed method consists of (1) a study-specific data-driven tractography parcellation to obtain white matter (WM) tract parcels according to the WM anatomy and (2) a nonparametric permutation-based STFC test to identify significant dif- ferences between study populations (e.g. disease and healthy). The basic idea of our method is that a WM parcel’s neighborhood (parcels with similar WM anatomy) can support the parcel’s statistical significance when correcting for multiple comparisons. The method is demonstrated by application to a multi-shell diffusion MRI dataset from 59 individuals, including 30 attention deficit hyperactivity disorder (ADHD) patients and 29 healthy controls (HCs). Evaluations are conducted using both synthetic and real data. The results indicate that our STFC method gives greater sensitivity in finding group differences in WM tract parcels compared to several traditional multiple comparison correction methods.
PURPOSE: Accurate segmentation of lung nodules is crucial in the development of imaging biomarkers for predicting malignancy of the nodules. Manual segmentation is time consuming and affected by inter-observer variability. We evaluated the robustness and accuracy of a publically available semiautomatic segmentation algorithm that is implemented in the 3D Slicer Chest Imaging Platform (CIP) and compared it with the performance of manual segmentation. METHODS: CT images of 354 manually segmented nodules were downloaded from the LIDC database. Four radiologists performed the manual segmentation and assessed various nodule characteristics. The semiautomatic CIP segmentation was initialized using the centroid of the manual segmentations, thereby generating four contours for each nodule. The robustness of both segmentation methods was assessed using the region of uncertainty (δ) and Dice similarity index (DSI). The robustness of the segmentation methods was compared using the Wilcoxon-signed rank test (pWilcoxon<0.05). The Dice similarity index (DSIAgree) between the manual and CIP segmentations was computed to estimate the accuracy of the semiautomatic contours. RESULTS: The median computational time of the CIP segmentation was 10 s. The median CIP and manually segmented volumes were 477 ml and 309 ml, respectively. CIP segmentations were significantly more robust than manual segmentations (median δCIP = 14ml, median dsiCIP = 99% vs. median δmanual = 222ml, median dsimanual = 82%) with pWilcoxon~10-16. The agreement between CIP and manual segmentations had a median DSIAgree of 60%. While 13% (47/354) of the nodules did not require any manual adjustment, minor to substantial manual adjustments were needed for 87% (305/354) of the nodules. CIP segmentations were observed to perform poorly (median DSIAgree≈50%) for non-/sub-solid nodules with subtle appearances and poorly defined boundaries. CONCLUSION: Semi-automatic CIP segmentation can potentially reduce the physician workload for 13% of nodules owing to its computational efficiency and superior stability compared to manual segmentation. Although manual adjustment is needed for many cases, CIP segmentation provides a preliminary contour for physicians as a starting point.