Florian Steinke

MRI-Based Attenuation Correction for PET/MRI: A Novel Approach Combining Pattern Recognition and Atlas Registration

For quantitative PET information, correction of tissue photon attenuation is mandatory. Generally in conventional PET, the attenuation map is obtained from a transmission scan, which uses a rotating radionuclide source, or from the CT scan in a combined PET/CT scanner. In the case of PET/MRI scanners currently under development, insufficient space for the rotating source exists; the attenuation map can be calculated from the MR image instead. This task is challenging because MR intensities correlate with proton densities and tissue-relaxation properties, rather than with attenuation-related mass density. Methods: We used a combination of local pattern recognition and atlas registration, which captures global variation of anatomy, to predict pseudo-CT images from a given MR image. These pseudo-CT images were then used for attenuation correction, as the process would be performed in a PET/CT scanner. Results: For human brain scans, we show on a database of 17 MR/CT image pairs that our method reliably enables estimation of a pseudo-CT image from the MR image alone. On additional datasets of MRI/PET/CT triplets of human brain scans, we compare MRI-based attenuation correction with CT-based correction. Our approach enables PET quantification with a mean error of 3.2% for predefined regions of interest, which we found to be clinically not significant. However, our method is not specific to brain imaging, and we show promising initial results on 1 whole-body animal dataset. Conclusion: This method allows reliable MRI-based attenuation correction for human brain scans. Further work is necessary to validate the method for whole-body imaging.

MRI-Based Attenuation Correction for Whole-Body PET/MRI: Quantitative Evaluation of Segmentation- and Atlas-Based Methods

PET/MRI is an emerging dual-modality imaging technology that requires new approaches to PET attenuation correction (AC). We assessed 2 algorithms for whole-body MRI-based AC (MRAC): a basic MR image segmentation algorithm and a method based on atlas registration and pattern recognition (AT&PR). Methods: Eleven patients each underwent a whole-body PET/CT study and a separate multibed whole-body MRI study. The MR image segmentation algorithm uses a combination of image thresholds, Dixon fat–water segmentation, and component analysis to detect the lungs. MR images are segmented into 5 tissue classes (not including bone), and each class is assigned a default linear attenuation value. The AT&PR algorithm uses a database of previously aligned pairs of MRI/CT image volumes. For each patient, these pairs are registered to the patient MRI volume, and machine-learning techniques are used to predict attenuation values on a continuous scale. MRAC methods are compared via the quantitative analysis of AC PET images using volumes of interest in normal organs and on lesions. We assume the PET/CT values after CT-based AC to be the reference standard. Results: In regions of normal physiologic uptake, the average error of the mean standardized uptake value was 14.1% ± 10.2% and 7.7% ± 8.4% for the segmentation and the AT&PR methods, respectively. Lesion-based errors were 7.5% ± 7.9% for the segmentation method and 5.7% ± 4.7% for the AT&PR method. Conclusion: The MRAC method using AT&PR provided better overall PET quantification accuracy than the basic MR image segmentation approach. This better quantification was due to the significantly reduced volume of errors made regarding volumes of interest within or near bones and the slightly reduced volume of errors made regarding areas outside the lungs.

Support Vector Machines for 3D Shape Processing

We propose statistical learning methods for approximating implicit surfaces and computing dense 3D deformation fields. Our approach is based on Support Vector (SV) Machines, which are state of the art in machine learning. It is straightforward to implement and computationally competitive; its parameters can be automatically set using standard machine learning methods. The surface approximation is based on a modified Support Vector regression. We present applications to 3D head reconstruction, including automatic removal of outliers and hole filling. In a second step, we build on our SV representation to compute dense 3D deformation fields between two objects. The fields are computed using a generalized SV Machine enforcing correspondence between the previously learned implicit SV object representations, as well as correspondences between feature points if such points are available. We apply the method to the morphing of 3D heads and other objects.

Experimental design for efficient identification of gene regulatory networks using sparse Bayesian models

BackgroundIdentifying large gene regulatory networks is an important task, while the acquisition of data through perturbation experiments (e.g., gene switches, RNAi, heterozygotes) is expensive. It is thus desirable to use an identification method that effectively incorporates available prior knowledge – such as sparse connectivity – and that allows to design experiments such that maximal information is gained from each one.ResultsOur main contributions are twofold: a method for consistent inference of network structure is provided, incorporating prior knowledge about sparse connectivity. The algorithm is time efficient and robust to violations of model assumptions. Moreover, we show how to use it for optimal experimental design, reducing the number of required experiments substantially. We employ sparse linear models, and show how to perform full Bayesian inference for these. We not only estimate a single maximum likelihood network, but compute a posterior distribution over networks, using a novel variant of the expectation propagation method. The representation of uncertainty enables us to do effective experimental design in a standard statistical setting: experiments are selected such that the experiments are maximally informative.ConclusionFew methods have addressed the design issue so far. Compared to the most well-known one, our method is more transparent, and is shown to perform qualitatively superior. In the former, hard and unrealistic constraints have to be placed on the network structure for mere computational tractability, while such are not required in our method. We demonstrate reconstruction and optimal experimental design capabilities on tasks generated from realistic non-linear network simulators.The methods described in the paper are available as a Matlab package athttp://www.kyb.tuebingen.mpg.de/sparselinearmodel.

Learning similarity measure for multi-modal 3D image registration

Multi-modal image registration is a challenging problem in medical imaging. The goal is to align anatomically identical structures; however, their appearance in images acquired with different imaging devices, such as CT or MR, may be very different. Registration algorithms generally deform one image, the floating image, such that it matches with a second, the reference image, by maximizing some similarity score between the deformed and the reference image. Instead of using a universal, but a priori fixed similarity criterion such as mutual information, we propose learning a similarity measure in a discriminative manner such that the reference and correctly deformed floating images receive high similarity scores. To this end, we develop an algorithm derived from max-margin structured output learning, and employ the learned similarity measure within a standard rigid registration algorithm. Compared to other approaches, our method adapts to the specific registration problem at hand and exploits correlations between neighboring pixels in the reference and the floating image. Empirical evaluation on CT-MR/PET-MR rigid registration tasks demonstrates that our approach yields robust performance and outperforms the state of the art methods for multi-modal medical image registration.

Kernels, regularization and differential equations

Many common machine learning methods such as support vector machines or Gaussian process inference make use of positive definite kernels, reproducing kernel Hilbert spaces, Gaussian processes, and regularization operators. In this work these objects are presented in a general, unifying framework and interrelations are highlighted. With this in mind we then show how linear stochastic differential equation models can be incorporated naturally into the kernel framework. And vice versa, many kernel machines can be interpreted in terms of differential equations. We focus especially on ordinary differential equations, also known as dynamical systems, and it is shown that standard kernel inference algorithms are equivalent to Kalman filter methods based on such models. In order not to cloud qualitative insights with heavy mathematical machinery, we restrict ourselves to finite domains, implying that differential equations are treated via their corresponding finite difference equations.

A novel metric for information retrieval in semantic networks

We propose a novel graph metric for semantic entity-relationship networks. The metric is used for solving two tasks. First, given a semantic entity-relationship graph, such as for example DBpedia, we find relevant neighbors for a given query node. This could be useful for retrieving information relating to a specific entity. Second, we search for paths between two given nodes to discover interesting links. As an example, this can be helpful to analyze the various relationships between Albert Einstein and Niels Bohr. Compared to using the default step metric our approach yields more specific and informative results, as we demonstrate using two semantic web datasets. The proposed metric is defined via paths that maximize the log-likelihood of a restricted round trip and can intuitively be interpreted in terms of random walks on graphs. Our distance metric is also related to the commute distance, which is highly plausible for the described tasks but prohibitively expensive to compute. Our metric can be calculated efficiently using standard graph algorithms, rendering the approach feasible for the very large graphs of the semantic webs linked data.

Semantic Traffic-Aware Routing Using the LarKC Platform

The popularity of location-based services and automotive navigation systems calls for a new generation of intelligent solutions to support users in mobility. This article presents a traffic-aware semantic routing service for mobile users based on the Large Knowledge Collider (LarKC) Semantic Web pluggable platform. It proposes a technique for integrating conceptual query answering with statistical learning and operations research algorithms. The presented prototype of a traffic-aware semantic routing service works efficiently with large, heterogeneous information sources and delivers value-added services to mobile users.

Nonparametric Regression between General Riemannian Manifolds

We study nonparametric regression between Riemannian manifolds based on regularized empirical risk minimization. Regularization functionals for mappings between manifolds should respect the geometry of input and output manifold and be independent of the chosen parametrization of the manifolds. We define and analyze the three most simple regularization functionals with these properties and present a rather general scheme for solving the resulting optimization problem. As application examples we discuss interpolation on the sphere, fingerprint processing, and correspondence computations between three-dimensional surfaces. We conclude with characterizing interesting and sometimes counterintuitive implications and new open problems that are specific to learning between Riemannian manifolds and are not encountered in multivariate regression in Euclidean space.

Scalable Economic Dispatch for Smart Distribution Networks

We present a novel algorithm for economic dispatch in electric power grids. The method is inspired by statistical inference methods. Using discretized optimization variables, our algorithm finds the globally optimal, single time-step dispatch assignment for radial grids in linear time with respect to the number of network nodes. For such problems, the algorithm outperforms state-of-the-art mixed-integer scheduling, both in run-time and in the allowed complexity of component and line models. Moreover, the necessary computations can be performed in a distributed fashion, facilitating both practical implementation as well as information privacy. Our algorithm is thus optimally suited for the very large dispatch problems that will arise in future smart distribution grids with hosts of small, decentralized, and flexibly controllable prosumers, i.e., entities able to consume and produce electricity.