banner research

Image Analysis

SCI's imaging work addresses fundamental questions in 2D and 3D image processing, including filtering, segmentation, surface reconstruction, and shape analysis. In low-level image processing, this effort has produce new nonparametric methods for modeling image statistics, which have resulted in better algorithms for denoising and reconstruction. Work with particle systems has led to new methods for visualizing and analyzing 3D surfaces. Our work in image processing also includes applications of advanced computing to 3D images, which has resulted in new parallel algorithms and real-time implementations on graphics processing units (GPUs). Application areas include medical image analysis, biological image processing, defense, environmental monitoring, and oil and gas.


Ross Whitaker


Sarang Joshi

Shape Statistics
Brain Atlasing

Tolga Tasdizen

Image Processing
Machine Learning

Tom Fletcher

Shape Statistics
Diffusion Tensor Analysis

Chris Johnson

Diffusion Tensor Analysis

Image Analysis Project Sites:

Publications in Image Analysis:

Segmentation of Serial MRI of TBI patients using Personalized Atlas Construction and Topological Change Estimation
Bo Wang, M.W. Prastawa, S.P. Awate, A. Irimia, M.C. Chambers, P.M. Vespa, J.D. Van Horn, G. Gerig. In Proceedings of IEEE ISBI 2012, pp. 1152--1155. 2012.
DOI: 10.1109/ISBI.2012.6235764

2D image space methods are processing methods applied after the volumetric data are projected and rendered into the 2D image space, such as 2D filtering, tone mapping and compositing. In the application domain of volume visualization, most 2D image space methods can be carried out more efficiently than their 3D counterparts. Most importantly, 2D image space methods can be used to enhance volume visualization quality when applied together with volume rendering methods. In this paper, we present and discuss the applications of a series of 2D image space methods as enhancements to confocal microscopy visualizations, including 2D tone mapping, 2D compositing, and 2D color mapping. These methods are easily integrated with our existing confocal visualization tool, FluoRender, and the outcome is a full-featured visualization system that meets neurobiologists' demands for qualitative analysis of confocal microscopy data.

Longitudinal Growth Modeling of Discrete-Time Functions with Application to DTI Tract Evolution in Early Neurodevelopment
A. Sharma, S. Durrleman, J.H. Gilmore, G. Gerig. In Proceedings of IEEE ISBI 2012, pp. 1397--1400. 2012.
DOI: 10.1109/ISBI.2012.6235829

We present a new framework for spatiotemporal analysis of parameterized functions attributed by properties of 4D longitudinal image data. Our driving application is the measurement of temporal change in white matter diffusivity of fiber tracts. A smooth temporal modeling of change from a discrete-time set of functions is obtained with an extension of the logistic growth model to time-dependent spline functions, capturing growth with only a few descriptive parameters. An unbiased template baseline function is also jointly estimated. Solution is demonstrated via energy minimization with an extension to simultaneous modeling of trajectories for multiple subjects. The new framework is validated with synthetic data and applied to longitudinal DTI from 15 infants. Interpretation of estimated model growth parameters is facilitated by visualization in the original coordinate space of fiber tracts.

Quantifying regional growth patterns through longitudinal analysis of distances between multimodal MR intensity distributions
A. Vardhan, M.W. Prastawa, S. Gouttard, J. Piven, G. Gerig. In Proceedings of IEEE ISBI 2012, pp. 1156--1159. 2012.
DOI: 10.1109/ISBI.2012.6235765

Quantitative analysis of early brain development through imaging is critical for identifying pathological development, which may in turn affect treatment procedures. We propose a framework for analyzing spatiotemporal patterns of brain maturation by quantifying intensity changes in longitudinal MR images. We use a measure of divergence between a pair of intensity distributions to study the changes that occur within specific regions, as well as between a pair of anatomical regions, over time. The change within a specific region is measured as the contrast between white matter and gray matter tissue belonging to that region. The change between a pair of regions is measured as the divergence between regional image appearances, summed over all tissue classes. We use kernel regression to integrate the temporal information across different subjects in a consistent manner. We applied our method on multimodal MRI data with T1-weighted (T1W) and T2-weighted (T2W) scans of each subject at the approximate ages of 6 months, 12 months, and 24 months. The results demonstrate that brain maturation begins at posterior regions and that frontal regions develop later, which matches previously published histological, qualitative and morphometric studies. Our multimodal analysis also confirms that T1W and T2W modalities capture different properties of the maturation process, a phenomena referred to as T2 time lag compared to T1. The proposed method has potential for analyzing regional growth patterns across different populations and for isolating specific critical maturation phases in different MR modalities.

A framework for longitudinal data analysis via shape regression
J. Fishbaugh, S. Durrleman, J. Piven, G. Gerig. In Medical Imaging 2012: Image Processing, Edited by David R. Haynor and Sebastien Ourselin, SPIE Intl Soc Optical Eng, Feb, 2012.
DOI: 10.1117/12.911721

Traditional longitudinal analysis begins by extracting desired clinical measurements, such as volume or head circumference, from discrete imaging data. Typically, the continuous evolution of a scalar measurement is estimated by choosing a 1D regression model, such as kernel regression or fitting a polynomial of fixed degree. This type of analysis not only leads to separate models for each measurement, but there is no clear anatomical or biological interpretation to aid in the selection of the appropriate paradigm. In this paper, we propose a consistent framework for the analysis of longitudinal data by estimating the continuous evolution of shape over time as twice differentiable flows of deformations. In contrast to 1D regression models, one model is chosen to realistically capture the growth of anatomical structures. From the continuous evolution of shape, we can simply extract any clinical measurements of interest. We demonstrate on real anatomical surfaces that volume extracted from a continuous shape evolution is consistent with a 1D regression performed on the discrete measurements. We further show how the visualization of shape progression can aid in the search for significant measurements. Finally, we present an example on a shape complex of the brain (left hemisphere, right hemisphere, cerebellum) that demonstrates a potential clinical application for our framework.

A Patient-Specific Segmentation Framework for Longitudinal MR Images of Traumatic Brain Injury
Bo Wang, M.W. Prastawa, A. Irimia, M.C. Chambers, P.M. Vespa, J.D. Van Horn, G. Gerig. In Proceedings of Medical Imaging 2012: Image Processing, SPIE, pp. 831402-831402-7. 2012.
DOI: 10.1117/12.911043

Traumatic brain injury (TBI) is a major cause of death and disability worldwide. Robust, reproducible segmentations of MR images with TBI are crucial for quantitative analysis of recovery and treatment efficacy. However, this is a significant challenge due to severe anatomy changes caused by edema (swelling), bleeding, tissue deformation, skull fracture, and other effects related to head injury. In this paper, we introduce a multi-modal image segmentation framework for longitudinal TBI images. The framework is initialized through manual input of primary lesion sites at each time point, which are then refined by a joint approach composed of Bayesian segmentation and construction of a personalized atlas. The personalized atlas construction estimates the average of the posteriors of the Bayesian segmentation at each time point and warps the average back to each time point to provide the updated priors for Bayesian segmentation. The difference between our approach and segmenting longitudinal images independently is that we use the information from all time points to improve the segmentations. Given a manual initialization, our framework automatically segments healthy structures (white matter, grey matter, cerebrospinal fluid) as well as different lesions such as hemorrhagic lesions and edema. Our framework can handle different sets of modalities at each time point, which provides flexibility in analyzing clinical scans. We show results on three subjects with acute baseline scans and chronic follow-up scans. The results demonstrate that joint analysis of all the points yields improved segmentation compared to independent analysis of the two time points.

Automatic corpus callosum segmentation using a deformable active Fourier contour model
C. Vachet, B. Yvernault, K. Bhatt, R.G. Smith, G. Gerig, H.C. Hazlett, M.A. Styner. In Proceedings of Medical Imaging 2012: Biomedical Applications in Molecular, Structural, and Functional Imaging, SPIE, Vol. 8317, 831707, 2012.
DOI: 10.1117/12.911504

The corpus callosum (CC) is a structure of interest in many neuroimaging studies of neuro-developmental pathology such as autism. It plays an integral role in relaying sensory, motor and cognitive information from homologous regions in both hemispheres.

We have developed a framework that allows automatic segmentation of the corpus callosum and its lobar subdivisions. Our approach employs constrained elastic deformation of exible Fourier contour model, and is an extension of Szekely's 2D Fourier descriptor based Active Shape Model. The shape and appearance model, derived from a large mixed population of 150+ subjects, is described with complex Fourier descriptors in a principal component shape space. Using MNI space aligned T1w MRI data, the CC segmentation is initialized on the mid-sagittal plane using the tissue segmentation. A multi-step optimization strategy, with two constrained steps and a final unconstrained step, is then applied. If needed, interactive segmentation can be performed via contour repulsion points. Lobar connectivity based parcellation of the corpus callosum can finally be computed via the use of a probabilistic CC subdivision model.

Our analysis framework has been integrated in an open-source, end-to-end application called CCSeg both with a command line and Qt-based graphical user interface (available on NITRC). A study has been performed to quantify the reliability of the semi-automatic segmentation on a small pediatric dataset. Using 5 subjects randomly segmented 3 times by two experts, the intra-class correlation coeficient showed a superb reliability (0.99). CCSeg is currently applied to a large longitudinal pediatric study of brain development in autism.

Differences in White Matter Fiber Tract Development Present from 6 to 24 Months in Infants with Autism,
J.J. Wolff, H. Gu, G. Gerig, J.T. Elison, M. Styner, S. Gouttard, K.N. Botteron, S.R. Dager, G. Dawson, A.M. Estes, A. Evans, H.C. Hazlett, P. Kostopoulos, R.C. McKinstry, S.J. Paterson, R.T. Schultz, L. Zwaigenbaum, J. Piven. In American Journal of Psychiatry (AJP), Note: Selected as an AJP Best of 2012 paper., pp. 1--12. 2012.
DOI: 10.1176/appi.ajp.2011.11091447

Objective: Evidence from prospective studies of high-risk infants suggests that early symptoms of autism usually emerge late in the first or early in the second year of life after a period of relatively typical development. The authors prospectively examined white matter fiber tract organization from 6 to 24 months in high-risk infants who developed autism spectrum disorders (ASDs) by 24 months.

Method: The participants were 92 highrisk infant siblings from an ongoing imaging study of autism. All participants had diffusion tensor imaging at 6 months and behavioral assessments at 24 months; a majority contributed additional imaging data at 12 and/or 24 months. At 24 months, 28 infants met criteria for ASDs and 64 infants did not. Microstructural properties of white matter fiber tracts reported to be associated with ASDs or related behaviors were characterized by fractional anisotropy and radial and axial diffusivity.

Results: The fractional anisotropy trajectories for 12 of 15 fiber tracts differed significantly between the infants who developed ASDs and those who did not. Development for most fiber tracts in the infants with ASDs was characterized by higher fractional anisotropy values at 6 months followed by slower change over time relative to infants without ASDs. Thus, by 24 months of age, those with ASDs had lower values.

Conclusions: These results suggest that aberrant development of white matter pathways may precede the manifestation of autistic symptoms in the first year of life. Longitudinal data are critical to characterizing the dynamic age-related brain and behavior changes underlying this neurodevelopmental disorder.

FluoRender: An Application of 2D Image Space Methods for 3D and 4D Confocal Microscopy Data Visualization in Neurobiology Research
Y. Wan, H. Otsuna, C.-B. Chien, C.D. Hansen. In Proceedings of Pacific Vis 2012, Incheon, Korea, pp. 201--208. 2012.
DOI: 10.1109/PacificVis.2012.6183592

2D image space methods are processing methods applied after the volumetric data are projected and rendered into the 2D image space, such as 2D filtering, tone mapping and compositing. In the application domain of volume visualization, most 2D image space methods can be carried out more efficiently than their 3D counterparts. Most importantly, 2D image space methods can be used to enhance volume visualization quality when applied together with volume rendering methods. In this paper, we present and discuss the applications of a series of 2D image space methods as enhancements to confocal microscopy visualizations, including 2D tone mapping, 2D compositing, and 2D color mapping. These methods are easily integrated with our existing confocal visualization tool, FluoRender, and the outcome is a full-featured visualization system that meets neurobiologists' demands for qualitative analysis of confocal microscopy data.

Keywords: scidac

Measures for Validation of DTI Tractography
S. Gouttard, C.B. Goodlett, M. Kubicki, G. Gerig. In Medical Imaging 2012: Image Processing, Edited by David R. Haynor and Sebastien Ourselin, SPIE Intl Soc Optical Eng, Feb, 2012.
DOI: 10.1117/12.911546

The evaluation of analysis methods for diffusion tensor imaging (DTI) remains challenging due to the lack of gold standards and validation frameworks. Significant work remains in developing metrics for comparing fiber bundles generated from streamline tractography. We propose a set of volumetric and tract oriented measures for evaluating tract differences. The different methods developed for this assessment work are: an overlap measurement, a point cloud distance and a quantification of the diffusion properties at similar locations between fiber bundles. The application of the measures in this paper is a comparison of atlas generated tractography to tractography generated in individual images. For the validation we used a database of 37 subject DTIs, and applied the measurements on five specific fiber bundles: uncinate, cingulum (left and right for both bundles) and genu. Each measurments is interesting for specific use: the overlap measure presents a simple and comprehensive metric but is sensitive to partial voluming and does not give consistent values depending on the bundle geometry. The point cloud distance associated with a quantile interpretation of the distribution gives a good intuition of how close and similar the bundles are. Finally, the functional difference is useful for a comparison of the diffusion properties since it is the focus of many DTI analysis to compare scalar invariants. The comparison demonstrated reasonable similarity of results. The tract difference measures are also applicable to comparison of tractography algorithms, quality control, reproducibility studies, and other validation problems.

Building Spatiotemporal Anatomical Models using Joint 4-D Segmentation, Registration, and Subject-Speci fic Atlas Estimation
M.W. Prastawa, S.P. Awate, G. Gerig. In Proceedings of the 2012 IEEE Mathematical Methods in Biomedical Image Analysis (MMBIA) Conference, pp. 49--56. 2012.
DOI: 10.1109/MMBIA.2012.6164740
PubMed ID: 23568185
PubMed Central ID: PMC3615562

Longitudinal analysis of anatomical changes is a vital component in many personalized-medicine applications for predicting disease onset, determining growth/atrophy patterns, evaluating disease progression, and monitoring recovery. Estimating anatomical changes in longitudinal studies, especially through magnetic resonance (MR) images, is challenging because of temporal variability in shape (e.g. from growth/atrophy) and appearance (e.g. due to imaging parameters and tissue properties affecting intensity contrast, or from scanner calibration). This paper proposes a novel mathematical framework for constructing subject-specific longitudinal anatomical models. The proposed method solves a generalized problem of joint segmentation, registration, and subject-specific atlas building, which involves not just two images, but an entire longitudinal image sequence. The proposed framework describes a novel approach that integrates fundamental principles that underpin methods for image segmentation, image registration, and atlas construction. This paper presents evaluation on simulated longitudinal data and on clinical longitudinal brain MRI data. The results demonstrate that the proposed framework effectively integrates information from 4-D spatiotemporal data to generate spatiotemporal models that allow analysis of anatomical changes over time.

Keywords: namic, adni, autism

Data Analysis with the Morse-Smale Complex: The msr Package for R
S. Gerber, K. Potter. In Journal of Statistical Software, Vol. 50, No. 2, 2012.

In many areas, scientists deal with increasingly high-dimensional data sets. An important aspect for these scientists is to gain a qualitative understanding of the process or system from which the data is gathered. Often, both input variables and an outcome are observed and the data can be characterized as a sample from a high-dimensional scalar function. This work presents the R package msr for exploratory data analysis of multivariate scalar functions based on the Morse-Smale complex. The Morse-Smale complex provides a topologically meaningful decomposition of the domain. The msr package implements a discrete approximation of the Morse-Smale complex for data sets. In previous work this approximation has been exploited for visualization and partition-based regression, which are both supported in the msr package. The visualization combines the Morse-Smale complex with dimension-reduction techniques for a visual summary representation that serves as a guide for interactive exploration of the high-dimensional function. In a similar fashion, the regression employs a combination of linear models based on the Morse-Smale decomposition of the domain. This regression approach yields topologically accurate estimates and facilitates interpretation of general trends and statistical comparisons between partitions. In this manner, the msr package supports high-dimensional data understanding and exploration through the Morse-Smale complex.

Centerline extraction with principal curve tracing to improve 3D level set esophagus segmentation in CT images
S. Kurugol, E. Bas, D. Erdogmus, J.G. Dy, G.C. Sharp, D.H. Brooks. In Proceedings of IEEE International Conference of the Engineering in Medicine and Biology Society (EMBS), pp. 3403--3406. 2011.
DOI: 10.1109/IEMBS.2011.6090921
PubMed ID: 2225507
PubMed Central ID: PMC3349355

For radiotherapy planning, contouring of target volume and healthy structures at risk in CT volumes is essential. To automate this process, one of the available segmentation techniques can be used for many thoracic organs except the esophagus, which is very hard to segment due to low contrast. In this work we propose to initialize our previously introduced model based 3D level set esophagus segmentation method with a principal curve tracing (PCT) algorithm, which we adapted to solve the esophagus centerline detection problem. To address challenges due to low intensity contrast, we enhanced the PCT algorithm by learning spatial and intensity priors from a small set of annotated CT volumes. To locate the esophageal wall, the model based 3D level set algorithm including a shape model that represents the variance of esophagus wall around the estimated centerline is utilized. Our results show improvement in esophagus segmentation when initialized by PCT compared to our previous work, where an ad hoc centerline initialization was performed. Unlike previous approaches, this work does not need a very large set of annotated training images and has similar performance.

Brain Maturation of Newborns and Infants
G. Gerig, J.H. Gilmore, W. Lin. Encyclopedia on Early Childhood Development (online): Brain Development in Children - According to Experts, Montreal, Quebec, Centre of Excellence for Early Childhood Development and Strategic Knowledge Cluster on Early Child Development, pp. 1--6. 2011.

Recently, imaging studies of early human development have received more attention, as improved modeling methods might lead to a clearer understanding of the origin, timing, and nature of differences in neurodevelopmental disorders. Non-invasive magnetic resonance imaging (MRI) can provide three-dimensional images of the infant brain in less than 20 minutes, with unprecedented anatomical details and contrast of brain anatomy, cortical and subcortical structures and brain connectivity. Repeating MRI at different stages of development, e.g., in yearly intervals starting after birth, gives scientists the opportunity to study the trajectory of brain growth and compare individual growth trajectories to normative models. These comparisons become highly relevant in personalized medicine, where early diagnosis is a critical juncture for timing and therapy types.

The Viking viewer for connectomics: scalable multi-user annotation and summarization of large volume data sets
J.R. Anderson, S. Mohammed, B.C. Grimm, B.W. Jones, P. Koshevoy, T. Tasdizen, R.T. Whitaker, R.E. Marc. In Journal of Microscopy, Vol. 241, No. 1, pp. 13--28. 2011.
DOI: 10.1111/j.1365-2818.2010.03402.x

Modern microscope automation permits the collection of vast amounts of continuous anatomical imagery in both two and three dimensions. These large data sets present significant challenges for data storage, access, viewing, annotation and analysis. The cost and overhead of collecting and storing the data can be extremely high. Large data sets quickly exceed an individual's capability for timely analysis and present challenges in efficiently applying transforms, if needed. Finally annotated anatomical data sets can represent a significant investment of resources and should be easily accessible to the scientific community. The Viking application was our solution created to view and annotate a 16.5 TB ultrastructural retinal connectome volume and we demonstrate its utility in reconstructing neural networks for a distinctive retinal amacrine cell class. Viking has several key features. (1) It works over the internet using HTTP and supports many concurrent users limited only by hardware. (2) It supports a multi-user, collaborative annotation strategy. (3) It cleanly demarcates viewing and analysis from data collection and hosting. (4) It is capable of applying transformations in real-time. (5) It has an easily extensible user interface, allowing addition of specialized modules without rewriting the viewer.

Keywords: Annotation, automated electron microscopy, citizen science, computational methods, connectome, networks, visualization

Semi-automated Neuron Boundary Detection and Slice Traversal Algorithm for Segmentation of Neurons from Electron Microscopy Images
E. Jurrus, S. Watanabe, R. Guily, A.R.C. Paiva, M.H. Ellisman, E.M. Jorgensen, T. Tasdizen. In Microscopic Image Analysis with Applications in Biology (MIAAB) Workshop, 2011.

Neuroscientists are developing new imaging techniques and generating large volumes of data in an effort to understand the complex structure of the nervous system. To aid in the analysis, new segmentation techniques for identifying neurons in these feature rich datasets are required. However, the extremely anisotropic resolution of the data makes segmentation and tracking across slices difficult. This paper presents a complete method for segmenting neurons in electron microscopy images and visualizing them in three dimensions. First, we present an advanced method for identifying neuron membranes, necessary for whole neuron segmentation, using a machine learning approach. Next, neurons are segmented in each two-dimensional section and connected using correlation of regions between sections. These techniques, combined with a visual user interface, enable users to quickly segment whole neurons in large volumes.

Microfluidic laminate-based phantom for diffusion tensor-magnetic resonance imaging
R. Samuel, H.J. Sant, F. Jiao, C.R. Johnson, B.K. Gale. In Journal of Micromech. Microeng., Vol. 21, pp. 095027--095038. 2011.
DOI: 10.1088/0960-1317/21/9/095027

Geometric Correspondence for Ensembles of Nonregular Shapes
M. Datar, Y. Gur, B. Paniagua, M. Styner, R.T. Whitaker. In Proceedings of Medical Image Computing and Computer-Assisted Intervention (MICCAI 2011), Lecture Notes in Computer Science (LNCS), Vol. 6892, pp. 368--375. 2011.
DOI: 10.1007/978-3-642-23629-7_45
PubMed ID: 21995050
PubMed Central ID: PMC3346950

An ensemble of biological shapes can be represented and analyzed with a dense set of point correspondences. In previous work, optimal point placement was determined by optimizing an information theoretic criterion that depends on relative spatial locations on different shapes combined with pairwise Euclidean distances between nearby points on the same shape. These choices have prevented such methods from effectively characterizing shapes with complex geometry such as thin or highly curved features. This paper extends previous methods for automatic shape correspondence by taking into account the underlying geometry of individual shapes. This is done by replacing the Euclidean distance for intrashape pairwise particle interactions by the geodesic distance. A novel set of numerical techniques for fast distance computations on curved surfaces is used to extract these distances. In addition, we introduce an intershape penalty term that incorporates surface normal information to achieve better particle correspondences near sharp features. Finally, we demonstrate this new method on synthetic and biological datasets.

Keywords: namic

Detection of crossing white matter fibers with high-order tensors and rank-k decompositions
F. Jiao, Y. Gur, C.R. Johnson, S. Joshi. In Proceedings of the International Conference on Information Processing in Medical Imaging (IPMI 2011), Lecture Notes in Computer Science (LNCS), Vol. 6801, pp. 538--549. 2011.
DOI: 10.1007/978-3-642-22092-0_44
PubMed Central ID: PMC3327305

Fundamental to high angular resolution diffusion imaging (HARDI), is the estimation of a positive-semidefinite orientation distribution function (ODF) and extracting the diffusion properties (e.g., fiber directions). In this work we show that these two goals can be achieved efficiently by using homogeneous polynomials to represent the ODF in the spherical deconvolution approach, as was proposed in the Cartesian Tensor-ODF (CT-ODF) formulation. Based on this formulation we first suggest an estimation method for positive-semidefinite ODF by solving a linear programming problem that does not require special parametrization of the ODF. We also propose a rank-k tensor decomposition, known as CP decomposition, to extract the fibers information from the estimated ODF. We show that this decomposition is superior to the fiber direction estimation via ODF maxima detection as it enables one to reach the full fiber separation resolution of the estimation technique. We assess the accuracy of this new framework by applying it to synthetic and experimentally obtained HARDI data.

Monte Carlo expectation maximization with hidden Markov models to detect functional networks in resting-state fMRI
W. Liu, S. Awate, J. Anderson, D. Yurgelun-Todd, P.T. Fletcher. In Machine Learning in Medical Imaging, Lecture Notes in Computer Science (LNCS), Vol. 7009/2011, pp. 59--66. 2011.
DOI: 10.1007/978-3-642-24319-6_8

Functional connectivity magnetic resonance imaging classification of autism
J.S. Anderson, J.A. Nielsen, A.L. Froehlich, M.B. DuBray, T.J. Druzgal, A.N. Cariello, J.R. Cooperrider, B.A. Zielinski, C. Ravichandran, P.T. Fletcher, A.L. Alexander, E.D. Bigler, N. Lange, J.E. Lainhart. In Brain, pp. (published online). 2011.
DOI: 10.1093/brain/awr263
PubMed ID: 22006979

Group differences in resting state functional magnetic resonance imaging connectivity between individuals with autism and typically developing controls have been widely replicated for a small number of discrete brain regions, yet the whole-brain distribution of connectivity abnormalities in autism is not well characterized. It is also unclear whether functional connectivity is sufficiently robust to be used as a diagnostic or prognostic metric in individual patients with autism. We obtained pairwise functional connectivity measurements from a lattice of 7266 regions of interest covering the entire grey matter (26.4 million connections) in a well-characterized set of 40 male adolescents and young adults with autism and 40 age-, sex- and IQ-matched typically developing subjects. A single resting state blood oxygen level-dependent scan of 8 min was used for the classification in each subject. A leave-one-out classifier successfully distinguished autism from control subjects with 83% sensitivity and 75\% specificity for a total accuracy of 79\% (P = 1.1 x 10-7). In subjects less than 20 years of age, the classifier performed at 89\% accuracy (P = 5.4 x 10-7). In a replication dataset consisting of 21 individuals from six families with both affected and unaffected siblings, the classifier performed at 71\% accuracy (91\% accuracy for subjects less than 20 years of age). Classification scores in subjects with autism were significantly correlated with the Social Responsiveness Scale (P = 0.05), verbal IQ (P = 0.02) and the Autism Diagnostic Observation Schedule-Generic's combined social and communication subscores (P = 0.05). An analysis of informative connections demonstrated that region of interest pairs with strongest correlation values were most abnormal in autism. Negatively correlated region of interest pairs showed higher correlation in autism (less anticorrelation), possibly representing weaker inhibitory connections, particularly for long connections (Euclidean distance greater than 10 cm). Brain regions showing greatest differences included regions of the default mode network, superior parietal lobule, fusiform gyrus and anterior insula. Overall, classification accuracy was better for younger subjects, with differences between autism and control subjects diminishing after 19 years of age. Classification scores of unaffected siblings of individuals with autism were more similar to those of the control subjects than to those of the subjects with autism. These findings indicate feasibility of a functional connectivity magnetic resonance imaging diagnostic assay for autism.