Publications

2015

Liu S, Liu S, Cai W, Pujol S, Kikinis R, Feng DD. Multi-Phase Feature Representation Learning for Neurodegenerative Disease Diagnosis. Artificial Life and Computational Intelligence. 2015;LNAI 8955:350–9.
Feature learning with high dimensional neuroimaging features has been explored for the applications on neurodegenerative diseases. Low-dimensional biomarkers, such as mental status test scores and cerebrospinal fluid level, are essential in clinical diagnosis of neurological disorders, because they could be simple and effective for the clinicians to assess the disorder’s progression and severity. Rather than only using the low-dimensional biomarkers as inputs for decision making systems, we believe that such low-dimensional biomarkers can be used for enhancing the feature learning pipeline. In this study, we proposed a novel feature representation learning framework, Multi-Phase Feature Representation (MPFR), with low-dimensional biomarkers embedded. MPFR learns high-level neuroimaging features by extracting the associations between the low-dimensional biomarkers and the highdimensional neuroimaging features with a deep neural network. We validated the proposed framework using the Mini-Mental-State-Examination (MMSE) scores as a low-dimensional biomarker and multi-modal neuroimaging data as the high-dimensional neuroimaging features from the ADNI baseline cohort. The proposed approach outperformed the original neural network in both binary and ternary Alzheimer’s disease classification tasks.
Lou Y, Tannenbaum A. Inter-modality Deformable Registration. Jia, X., and Jiang, S.B. (Eds.). (2015). Graphics Processing Unit-Based High Performance Computing in Radiation Therapy. CRC Press; 2015.
Deformable image registration (DIR) is one of the major problems in medical image processing, such as dose calculation [18], treatment planning [33] and scatter removal of cone beam CT (CBCT) [22]. It is of prime importance to establish a pixel-to-pixel correspondence between two images in many clinical scenarios. For instance, registration of a CT image to MRI of a patient taken at different time can provide complementary diagnostic information. For applications as such, since the deformation of the patient anatomy cannot be represented by a rigid transform, DIR is almost the sole means to establish this mapping. DIR can be generally categorized into intra-modality and inter-modality, or multi-modality. While intra-modality DIR can be easily handled by conventional intensity-based methods [11, 30], intermodality DIR problems are still far from being satisfactory. Yet, since different imaging modalities usually provide their unique angles to reveal patient anatomy and delineate microscopic disease, intermodality registration plays a key role to combine the information from multiple modalities to facilitate diagnostics and treatment of a certain disease.
Lou Y, Tannenbaum A. Inter-modality Deformable Registration. Jia, X., and Jiang, S.B. (Eds.). (2015). Graphics Processing Unit-Based High Performance Computing in Radiation Therapy. CRC Press; 2015.
Deformable image registration (DIR) is one of the major problems in medical image processing, such as dose calculation [18], treatment planning [33] and scatter removal of cone beam CT (CBCT) [22]. It is of prime importance to establish a pixel-to-pixel correspondence between two images in many clinical scenarios. For instance, registration of a CT image to MRI of a patient taken at different time can provide complementary diagnostic information. For applications as such, since the deformation of the patient anatomy cannot be represented by a rigid transform, DIR is almost the sole means to establish this mapping. DIR can be generally categorized into intra-modality and inter-modality, or multi-modality. While intra-modality DIR can be easily handled by conventional intensity-based methods [11, 30], intermodality DIR problems are still far from being satisfactory. Yet, since different imaging modalities usually provide their unique angles to reveal patient anatomy and delineate microscopic disease, intermodality registration plays a key role to combine the information from multiple modalities to facilitate diagnostics and treatment of a certain disease.

2014

Wachinger C, Golland P, Reuter M, Wells WM III. Gaussian Process Interpolation for Uncertainty Estimation in Image Registration. Med Image Comput Comput Assist Interv. 2014;17(Pt 1):267–74.

Intensity-based image registration requires resampling images on a common grid to evaluate the similarity function. The uncertainty of interpolation varies across the image, depending on the location of resampled points relative to the base grid. We propose to perform Bayesian inference with Gaussian processes, where the covariance matrix of the Gaussian process posterior distribution estimates the uncertainty in interpolation. The Gaussian process replaces a single image with a distribution over images that we integrate into a generative model for registration. Marginalization over resampled images leads to a new similarity measure that includes the uncertainty of the interpolation. We demonstrate that our approach increases the registration accuracy and propose an efficient approximation scheme that enables seamless integration with existing registration methods.

Wachinger C, Golland P, Reuter M. BrainPrint: Identifying Subjects by Their Brain. Med Image Comput Comput Assist Interv. 2014;17(Pt 3):41–8.

Introducing BrainPrint, a compact and discriminative representation of anatomical structures in the brain. BrainPrint captures shape information of an ensemble of cortical and subcortical structures by solving the 2D and 3D Laplace-Beltrami operator on triangular (boundary) and tetrahedral (volumetric) meshes. We derive a robust classifier for this representation that identifies the subject in a new scan, based on a database of brain scans. In an example dataset containing over 3000 MRI scans, we show that BrainPrint captures unique information about the subject’s anatomy and permits to correctly classify a scan with an accuracy of over 99.8%. All processing steps for obtaining the compact representation are fully automated making this processing framework particularly attractive for handling large datasets.

Batmanghelich KN, Cho M, San Jose R, Golland P. Spherical Topic Models for Imaging Phenotype Discovery in Genetic Studies. Bayesian Graph Models Biomed Imaging. 2014;8677:107–17.

In this paper, we use Spherical Topic Models to discover the latent structure of lung disease. This method can be widely employed when a measurement for each subject is provided as a normalized histogram of relevant features. In this paper, the resulting descriptors are used as phenotypes to identify genetic markers associated with the Chronic Obstructive Pulmonary Disease (COPD). Features extracted from images capture the heterogeneity of the disease and therefore promise to improve detection of relevant genetic variants in Genome Wide Association Studies (GWAS). Our generative model is based on normalized histograms of image intensity of each subject and it can be readily extended to other forms of features as long as they are provided as normalized histograms. The resulting algorithm represents the intensity distribution as a combination of meaningful latent factors and mixing co-efficients that can be used for genetic association analysis. This approach is motivated by a clinical hypothesis that COPD symptoms are caused by multiple coexisting disease processes. Our experiments show that the new features enhance the previously detected signal on chromosome 15 with respect to standard respiratory and imaging measurements.

Wachinger C, Golland P. Atlas-based Under-segmentation. Med Image Comput Comput Assist Interv. 2014;17(Pt 1):315–22.

We study the widespread, but rarely discussed, tendency of atlas-based segmentation to under-segment the organs of interest. Commonly used error measures do not distinguish between under- and over-segmentation, contributing to the problem. We explicitly quantify over- and under-segmentation in several typical examples and present a new hypothesis for the cause. We provide evidence that segmenting only one organ of interest and merging all surrounding structures into one label creates bias towards background in the label estimates suggested by the atlas. We propose a generative model that corrects for this effect by learning the background structures from the data. Inference in the model separates the background into distinct structures and consequently improves the segmentation accuracy. Our experiments demonstrate a clear improvement in several applications.

Rathi Y, Ning L, Michailovich O, Liao H, Gagoski B, Grant E, Shenton ME, Stern R, Westin CF, Lin A. Maximum Entropy Estimation of Glutamate and Glutamine in MR Spectroscopic Imaging. Med Image Comput Comput Assist Interv. 2014;17(Pt 2):749–56.

Magnetic resonance spectroscopic imaging (MRSI) is often used to estimate the concentration of several brain metabolites. Abnormalities in these concentrations can indicate specific pathology, which can be quite useful in understanding the disease mechanism underlying those changes. Due to higher concentration, metabolites such as N-acetylaspartate (NAA), Creatine (Cr) and Choline (Cho) can be readily estimated using standard Fourier transform techniques. However, metabolites such as Glutamate (Glu) and Glutamine (Gln) occur in significantly lower concentrations and their resonance peaks are very close to each other making it difficult to accurately estimate their concentrations (separately). In this work, we propose to use the theory of ’Spectral Zooming’ or high-resolution spectral analysis to separate the Glutamate and Glutamine peaks and accurately estimate their concentrations. The method works by estimating a unique power spectral density, which corresponds to the maximum entropy solution of a zero-mean stationary Gaussian process. We demonstrate our estimation technique on several physical phantom data sets as well as on in-vivo brain spectroscopic imaging data. The proposed technique is quite general and can be used to estimate the concentration of any other metabolite of interest.

Dalca AV, Sridharan R, Cloonan L, Fitzpatrick KM, Kanakis A, Furie KL, Rosand J, Wu O, Sabuncu M, Rost NS, Golland P. Segmentation 0f Cerebrovascular Pathologies in Stroke Patients with Spatial and Shape Priors. Med Image Comput Comput Assist Interv. 2014;17(Pt 2):773–80.

We propose and demonstrate an inference algorithm for the automatic segmentation of cerebrovascular pathologies in clinical MR images of the brain. Identifying and differentiating pathologies is important for understanding the underlying mechanisms and clinical outcomes of cerebral ischemia. Manual delineation of separate pathologies is infeasible in large studies of stroke that include thousands of patients. Unlike normal brain tissues and structures, the location and shape of the lesions vary across patients, presenting serious challenges for prior-driven segmentation. Our generative model captures spatial patterns and intensity properties associated with different cerebrovascular pathologies in stroke patients. We demonstrate the resulting segmentation algorithm on clinical images of a stroke patient cohort.