Demons Methods for Digital Mammography Registration

Size: px
Start display at page:

Download "Demons Methods for Digital Mammography Registration"

Transcription

1 Demons Methods for Digital Mammography Registration Yago Díez 1, Meritxell Tortajada 1,SergiGanau 2, Lidia Tortajada 2, Melcior Sentís 2, and Robert Martí 1 1 Computer Vision and Robotics Group, University of Girona, Spain {yago,txell,marly}@eia.udg.edu 2 UDIAT-CD, Corporació Sanitària Parc taulí, Sabadell, Spain Abstract. Image registration is regarded as an important tool for the analysis of temporal mammograms, specially for lesion detection. However, it has not been widely adopted due to the complexity of the problem and the limitations of the algorithms. This work evaluates the suitability of Diffeomorphic Demons for temporal mammographic registration. Analysis in a dataset of 440 images from 53 different patients (performing over 3300 registrations) shows a better registration quality using different objective measures compared to standard Rigid and Non-Rigid algorithms. In addition, a study of the effects of lesions in the registration is performed, indicating that registration results (specially for the Demons) can be potentially used for lesion detection. 1 Introduction Medical imaging is one of the most commonly used application areas for image registration, specially for non-rigid paradigms (see for instance the survey in [1]). The analysis of mammographic images is not an exception [2,6]. Most of the published approaches focus on registering image features such as breast boundary or internal regions such as, for example, the pectoral muscle. However, a reduced number of approaches have been proposed using intensity based information only. Moreover, it was reported in [3] that Rigid transformations obtained more robust results compared to Non-Rigid methods in terms of producing less nonrealistic distortions. More recently, the work in [4] where authors evaluate a larger number of Rigid and Non-Rigid intensity based algorithms, shows that although these unrealistic deformations can occur, well parametrized Non-Rigid algorithms obtain a better overall performance. The main goal of this work is to evaluate the suitability of a particular implementation of the Demons algorithm, the Diffeomorphic Demons, for temporal mammographic registration compared to other existing methods. Authors have shown [4] that the conventional Demons algorithm, although obtaining satisfactory results for similarity metrics, is deemed unsuitable for mammography analysis by perceptual studies due to the severe distortions induced. Complementary goals of this research are also the proposal of new measures for evaluating the quality of registration and the study of the effects of breast lesions (masses and J.M. Sanches, L. Micó, and J.S. Cardoso (Eds.): IbPRIA 2013, LNCS 7887, pp , c Springer-Verlag Berlin Heidelberg 2013

2 254 Y. Díez et al. architectural distortions) in the registration of a large database of patients with temporal information (images from the same breast acquired at different times). This paper is structured as follows: in section 2, we briefly describe the image registration algorithms and their implementation. Subsequently, registration results are presented, providing details on the data, experiments and quantitative analysis (Sec. 3). Finally, discussions and conclusions are provided (Sec. 4). 2 Materials and Methods 2.1 Data Base The images used in this study were obtained from our local digital database. We used 440 full-field digital (FFD) mammograms from 53 patients with breast cancer who had at least two mammographic studies, specifically 49 patients have two studies and 4 patients have three studies. Each study contains four mammographic images, two views (cranio-caudal and medio-lateral oblique) of left and right breasts. Mammograms were acquired using a Selenia FFD mammography system, with resolution 70 micron per pixel, size 4096x3328 or 2560x3328, and 12-bit depth. This work is focused on temporal comparison of mammograms, therefore each mammographic image is registered with its homonymous mammogram in posteriors studies using different registration methodologies doing in total more than 3300 temporal registrations. Presence of masses was annotated by expert radiologists and this allowed us to distinguish between those registration instances containing masses and those not containing them. 2.2 Image Preprocessing During the preprocessing stage, the breast skin-line and the pectoral muscle were manually segmented. Background and pectoral muscle areas were eliminated to preserve the breast area. To reduce computation time, images were resized (quarter size image) using bilinear interpolation. In addition, mammograms were flipped when necessary to match the orientation of the breasts in each mammogram. After that, a peripheral enhancement was applied to compensate thickness variations in breast periphery. During mammographic acquisitions, breast is compressed with a tilting compression paddle, so breast thickness can be non uniform, being lower in the periphery and overexposing this area. The peripheral enhancement determines the overexposed area by Otsu s thresholding and applies a correction factor over the detected region. Each pixel is divided by the mean value of its real neighbourhood and multiplied by the mean value of its ideal neighbourhood according continuity constraints [12]. 2.3 Registration Methods The Diffeomorphic Demons [13] algorithm derives from Thyrion s Demons [11]. The authors reformulated the original idea by formalizing Demons optimization

3 Demons Methods for Digital Mammography Registration 255 over the space of displacement fields. The method has been since then used successfully in a variety of registration scenarios including brain magnetic resonance imaging registration [9]. Furthermore, the authors provided different variants corresponding to the operation allowed in the space of deformation field: exponentiation, addition and composition. In this work, we will focus on the additive variant. For comparison purposes, we also considered two widely used methods for mammography registration: Affine and B-splines [4]. Furthermore, we use all these methods to study the role of some commonly used registration aspects such as Affine initialization and multi-resolution (MR). From now on, Additive Difeomorphic Demons with MR but without Affine initialization is noted md. Additive Diffeomorphic Demons without MR but with Affine initialization is noted AD and Additive Diffeomorphic Demons with MR using also MR Affine initialization is noted mamd. As for Non-Demons methods, A stands for plain Affine registration, mb corresponds to MR B-splines without Affine initialization, AB represent single resolution B-splines with Affine initialization and mamb includes both MR Affine initialization and MR B-splines. 2.4 Statistical Analysis We use boxplots as a compact way to describe thousands of data. Specifically, throughout section 3 we will present several multiple boxplots where each box will group the data resulting from a registration method. Although this is a convenient way to visualize data, statistical inference is necessary in order to provide objective backup to observations. As in this paper we compare the performance of several methods. Pairwise t-tests are the natural option, but given the high number of method pairs, their use is impractical. We present selected t-tests to highlight specific comparisons. In order to present pairwise comparisons between methods in a compact way, we use permutation tests. These tests choose pairs of methods and small sets of independent values obtained by them. Then pairwise hypothesis tests are performed. These tests are regarded in [10] to compute the p-value more exactly than usual t-tests. Finally, the number of times when p<0.05 is stored for every method. All these steps are repeated and what we present is the mean and standard deviation (μ, σ) of the times when each method produced significant p-values. Consequently, methods with higher means have passed a higher number of pairwise comparisons using randomly chosen subsets of values. Following [9], methods are presented in ranks determined by the best performing method (with mean and standard deviation noted μ 0,σ 0 ). Ranks are decided in terms of the distance of method means to μ 0.Specifically,rank1 methods are those in (μ 0 σ 0,μ 0 ], rank 2 methods fall in (μ 0 2σ 0,μ 0 σ 0 ]and finally rank 3 methods are those in the interval (μ 0 3σ 0,μ 0 2σ 0 ]. See [9,10] for further details on permutation tests.

4 256 Y. Díez et al. 3 Experimental Study All registration methods were implemented using the Insight Toolkit (itk) libraries [5]. We used 128 histogram bins and samples for metric computations. A minimum step length stopping criteria was also used. For practical reasons we also fixed a maximum number of iterations for all methods to a maximum of 1000 iterations. In combined or MR methods these iterations were evenly distributed between the methods or MR levels. The Diffeomorphic Demons implementation used for the calculations in this paper can be downloaded at [7]. 3.1 Metrics Metric values are the most widely used tool to measure image similarity in computer vision. Concerning their application to intensity-based registration of medical images, their use is twofold: first they provide the aforementioned quantification of similarity between images. Second, the registration process in usually formalized as an optimization problem and metric values determine both stopping conditions and solution update. This is true for B-splines and Affine registration but not for Diffeomorphic Demons, as these methods use an optimization function defined over the space of displacement fields. Consequently, metric measurements provide an objective value on how successful some methods were in terms of optimization and how similar two images are in their own particular terms. Unfortunately, a metric that is able to express exactly what medical experts perceive as a better registration does not yet exist, so other criteria are also necessary. In the results presented, the Mutual Information metric (MI) was used as the function to optimize for Affine and B- splines registration. Data on the Sum of Squared Distances metric (SSD) is also provided in order to gain insight in the two mentioned aspects of the process. Table 1 presents the permutation tests for the two metrics studied: MI and SSD. As is usual with permutation test, only methods that achieve positive test means are presented. The data shows how for both metrics the methods able to achieve the highest positive test means are Demons based methods. The best method in this pairwise comparison is always mamd (combining Affine initialization and MR Demons), other Demons-based methods follow, but their ranks are reversed depending on which metric is considered. Table 1. Permutation test for MI and SSD MI SSD METHOD μ σ METHOD μ σ Rank 1 mamd mamd md AD Rank 2 AD md Rank 3 AB AB

5 A Demons Methods for Digital Mammography Registration 257 The mentioned trends are much more evident when presented visually in terms of metric boxplots. Fig. 1 provides compared performances of Demons and B-splines based methods as well as standalone Affine registration. Data is provided separately for registrations with and without masses. For the sake of concreteness, all data corresponds to MI, although similar trends are observed for SSD. Notice how, the mean values presented in table 1 and figure 1 are not directly related. In the first case, the permutation test deal with how many times a particular method has passed a pairwise hypothesis test against another method while in the second case, the mean MI value presented relates to the degree of success obtained by each method after registration. We observe how Demons based methods do better than other approaches. We also see how combining both Affine initialization and MR works best for Demons and B-splines. If only one of the two improvements is used, Demons methods seem to benefit more from MR while B-splines favour Affine initialization. Concerning the presence of masses, mean values for every method slightly decrease. This is clearer for Demons methods and suggests the possibility of using Demons methods in registration applications such as lesion or mass detection. This possibility will be addressed further in section A BEF md AD (a) mamd mb AB mamb BEF md AD (b) mamd mb AB mamb Fig. 1. Mutual information. Temporal registrations a) without masses and b) with masses. Measurements before registration are noted as BEF. Values correspond to MI, so higher positive values stand for better results. 3.2 Deformation Fields In Non-Rigid registration, every pixel is transformed in a way that does not necessarily be related to its neighbouring pixels. Each of these movements can be expressed by a displacement vector. The union of all these vectors stands for a displacement (or deformation) field that characterises the Non-Rigid transformation. Fig. 2 shows a visual example of a deformation field. For ease of

6 258 Y. Díez et al. (a) (b) (c) Fig. 2. Deformation field example: (a) source image, (b) target image with visible mass and (c) corresponding AD deformation field visualization, only vector norms are presented in the image in the form of a color code. Pixels closer to red correspond to higher norm values and pixels closer to white to lower values. The figure shows the potential of deformation fields for applications such as mass detection. Pixels corresponding to the mass appearing in the second temporal study present the highest norm values in the whole of the image. We study deformation fields corresponding to Demons and B-spline methods. For the automatic study of these fields, several parameters can be considered [8]. We focus on two of them: average and maximum norms of the vectors in the deformation field. Our aim is to provide statistical support to the claim that flexible registration methods, specially Diffeomorphic Demons, can be used for mass detection in digital mammography images. Concerning the average norm of the deformation field of the methods studied, we computed the mean values of this average norm for every method before and after registration. A noticeable increase was observed, specially for Diffeomorphic Demons methods. This behaviour was observed even more clearly for the maximumnormindicator(seetable2). In order to provide inferential backing to these observations of descriptive statistics, a set of pairwise t-tests was performed to check if the maximum norm of a given method was significantly different for registrations with and without masses. The alternative hypothesis (H 1 ) was stated as the maximum norm is higher for registrations with masses than for those without masses. Results showed statistically significant difference for md (p value = ) and for mamd (p value = ). This did not, however, happen, for any of the B-splines methods whose p-values were all over 0.4. This shows the potential of Diffeomorphic Demons methods for mass detection in digital mammography. Results of the maximum norm of the vectors in the deformation field are also visualised in Fig 3. Maximum norms for all Non-Rigid methods studied are presented, data is provided separately for registrations with and without masses.

7 Demons Methods for Digital Mammography Registration 259 Table 2. Statistical summary of deformation fields for all methods. Registrations with and without masses are presented separately. Average Norm Max Norm Without Masses With Masses Without Masses With Masses μ σ μ σ μ σ μ σ md AD mamd mb AB mamb We observe how Demons methods produce deformations with higher maximum norm than B-splines. We also see how, as expected, using Affine initialization decreases the maximum norm. This happens as the initial Affine step helps cover part of the distance that each pixel has to cover in order to reach its corresponding pixel. Furthermore, the increase in maximum norm for registrations with masses is also visible md AD mamd (a) mb AB mamb md AD mamd (b) mb AB mamb Fig. 3. Deformation Field Maximum Norm. Temporal registrations a) without masses and b) with masses. 4 Discussion and Conclusions We have presented results using over 3300 registrations corresponding to breast cancer patients. The analysis of metric data shows how Demons based methods outperform Affine and B-splines based approaches. We have also studied the use of MR and Affine initialization for flexible methods showing how the best results are obtained by the combination of the two techniques. Access to annotations

8 260 Y. Díez et al. from expert radiologist allowed us to discriminate those registrations containing masses from those not containing them. The analysis of differences between both scenarios using deformation fields showed statistically significant differences in behaviour for Demons methods. To sum up, according to our study, not only does Demons provide best results in terms of measures commonly used to evaluate the quality of registration but it also shows application possibilities in specific fieldssuchasmassdetection. Acknowledgment. This work was partially supported by the Spanish Science and Innovation grant TIN C References 1. Zitova, B.: Image registration methods: a survey. Image and Vision Computing 21(11), (2003) 2. Guo, Y., Sivaramakrishna, R., Lu, C.C., Suri, J., Laxminarayan, S.: Breast image registration techniques: a survey. Medical and Biological Engineering and Computing 44(1), (2006) 3. van Engeland, S., Snoeren, P., Hendriks, J., Karssemeijer, N.: A comparison of methods for mammogram registration. IEEE Transactions on Medical Imaging 22(11), (2003) 4. Díez, Y., Oliver, A., Lladó, X., Freixenet, J., Martí, J., Vilanova, J., Martí, R.: Revisiting intensity-based image registration applied to mammography. IEEE Transactions on Information Technology in Biomedicine 15(5), (2011) 5. Ibáñez, L., Schroeder, W., Ng, L., Cates, J.: The ITK Software Guide, 1st edn. Kitware, Inc. (2003), ISBN Díez, Y., Oliver, A., Lladó, X., Martí, R.: Comparison of registration methods using mammographic images. In: IEEE International Conference on Image Processing, pp (2010) 7. Vercauteren, T., Pennec, X., Perchant, A., Ayache, N.: Diffeomorphic Demons Using ITK s Finite Difference Solver Hierarchy, 8. Janssens, G., Jacques, L., Orban de Xivry, J., Geets, X., Macq, B.: Diffeomorphic registration of images with variable contrast enhancement. International Journal of Biomedical Imaging 2011(891585), 16 pages, doi: /2011/ Klein, A., Andersson, J., Ardekani, B.A., Ashburner, J., Avants, B., Chiang, M., Christensen, G.E., Collins, D.L., Gee, J., Hellier, P., Hyun Song, J., Jenkinson, M., Lepage, C., Rueckert, D., Thompson, P., Vercauteren, T., Woods, R.P., Mann, J.J., Parseya, R.V.: Evaluation of 14 nonlinear deformation algorithms applied to human brain MRI registration. NeuroImage 46(3), (2009) 10. Menke, J., Martinez, T.: Using permutations instead of Student s t distribution for p-values in paired difference algorithm comparisons. In: Proceedings of IEEE International Joint Conference on Neural Networks, pp (2004) 11. Thirion, J.: Fast non-rigid matching of 3D medical images. HAL-CCSd-CNR (1995) 12. Tortajada, M., Oliver, A., Martí, R., Vilagran, M., Ganau, S., Tortajada, L., Sentís, M., Freixenet, J.: Adapting breast density classification from digitized to full-field digital mammograms. In: Maidment, A.D.A., Bakic, P.R., Gavenonis, S. (eds.) IWDM LNCS, vol. 7361, pp Springer, Heidelberg (2012) 13. Vercauteren, T., Pennec, X., Perchant, A., Ayache, N.: Diffeomorphic demons: efficient non-parametric image registration. NeuroImage 45(1-1) (2009)

Atlas Based Segmentation of the prostate in MR images

Atlas Based Segmentation of the prostate in MR images Atlas Based Segmentation of the prostate in MR images Albert Gubern-Merida and Robert Marti Universitat de Girona, Computer Vision and Robotics Group, Girona, Spain {agubern,marly}@eia.udg.edu Abstract.

More information

Automatic Seed Placement for Breast Lesion Segmentation on US Images

Automatic Seed Placement for Breast Lesion Segmentation on US Images Automatic Seed Placement for Breast Lesion Segmentation on US Images Joan Massich 1,, Fabrice Meriaudeau 2, Melcior Sentís 3,SergiGanau 3, Elsa Pérez 4, Robert Martí 1, Arnau Oliver 1, and Joan Martí 1

More information

Automatic Registration-Based Segmentation for Neonatal Brains Using ANTs and Atropos

Automatic Registration-Based Segmentation for Neonatal Brains Using ANTs and Atropos Automatic Registration-Based Segmentation for Neonatal Brains Using ANTs and Atropos Jue Wu and Brian Avants Penn Image Computing and Science Lab, University of Pennsylvania, Philadelphia, USA Abstract.

More information

Nonrigid Registration using Free-Form Deformations

Nonrigid Registration using Free-Form Deformations Nonrigid Registration using Free-Form Deformations Hongchang Peng April 20th Paper Presented: Rueckert et al., TMI 1999: Nonrigid registration using freeform deformations: Application to breast MR images

More information

Segmentation of the Pectoral Muscle in Breast MRI Using Atlas-Based Approaches

Segmentation of the Pectoral Muscle in Breast MRI Using Atlas-Based Approaches Segmentation of the Pectoral Muscle in Breast MRI Using Atlas-Based Approaches Albert Gubern-Mérida 1, Michiel Kallenberg 2, Robert Martí 1, and Nico Karssemeijer 2 1 University of Girona, Spain {agubern,marly}@eia.udg.edu

More information

Whole Body MRI Intensity Standardization

Whole Body MRI Intensity Standardization Whole Body MRI Intensity Standardization Florian Jäger 1, László Nyúl 1, Bernd Frericks 2, Frank Wacker 2 and Joachim Hornegger 1 1 Institute of Pattern Recognition, University of Erlangen, {jaeger,nyul,hornegger}@informatik.uni-erlangen.de

More information

Breast Image Registration Using Non-Linear Local Affine Transformation

Breast Image Registration Using Non-Linear Local Affine Transformation Breast Image Registration Using Non-Linear Local Affine Transformation Feiyu Chen a, Peng Zheng a, Penglong Xu a, Andrew D. A. Maidment b, Predrag R. Bakic b, David D. Pokrajac a, Fengshan Liu a, Xiquan

More information

Non-Rigid Registration of Medical Images: Theory, Methods and Applications

Non-Rigid Registration of Medical Images: Theory, Methods and Applications Non-Rigid Registration of Medical Images: Theory, Methods and Applications Daniel Rueckert Paul Aljabar Medical mage registration [1] plays an increasingly important role in many clinical applications

More information

Learning-based Neuroimage Registration

Learning-based Neuroimage Registration Learning-based Neuroimage Registration Leonid Teverovskiy and Yanxi Liu 1 October 2004 CMU-CALD-04-108, CMU-RI-TR-04-59 School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213 Abstract

More information

Non-Rigid Registration of Medical Images: Theory, Methods and Applications

Non-Rigid Registration of Medical Images: Theory, Methods and Applications Non-Rigid Registration of Medical Images: Theory, Methods and Applications Daniel Rueckert Paul Aljabar Medical image registration [1] plays an increasingly important role in many clinical applications

More information

Ensemble registration: Combining groupwise registration and segmentation

Ensemble registration: Combining groupwise registration and segmentation PURWANI, COOTES, TWINING: ENSEMBLE REGISTRATION 1 Ensemble registration: Combining groupwise registration and segmentation Sri Purwani 1,2 sri.purwani@postgrad.manchester.ac.uk Tim Cootes 1 t.cootes@manchester.ac.uk

More information

Methods for data preprocessing

Methods for data preprocessing Methods for data preprocessing John Ashburner Wellcome Trust Centre for Neuroimaging, 12 Queen Square, London, UK. Overview Voxel-Based Morphometry Morphometry in general Volumetrics VBM preprocessing

More information

Automatic Subthalamic Nucleus Targeting for Deep Brain Stimulation. A Validation Study

Automatic Subthalamic Nucleus Targeting for Deep Brain Stimulation. A Validation Study Automatic Subthalamic Nucleus Targeting for Deep Brain Stimulation. A Validation Study F. Javier Sánchez Castro a, Claudio Pollo a,b, Jean-Guy Villemure b, Jean-Philippe Thiran a a École Polytechnique

More information

Automatic Vascular Tree Formation Using the Mahalanobis Distance

Automatic Vascular Tree Formation Using the Mahalanobis Distance Automatic Vascular Tree Formation Using the Mahalanobis Distance Julien Jomier, Vincent LeDigarcher, and Stephen R. Aylward Computer-Aided Diagnosis and Display Lab, Department of Radiology The University

More information

PROSTATE CANCER DETECTION USING LABEL IMAGE CONSTRAINED MULTIATLAS SELECTION

PROSTATE CANCER DETECTION USING LABEL IMAGE CONSTRAINED MULTIATLAS SELECTION PROSTATE CANCER DETECTION USING LABEL IMAGE CONSTRAINED MULTIATLAS SELECTION Ms. Vaibhavi Nandkumar Jagtap 1, Mr. Santosh D. Kale 2 1 PG Scholar, 2 Assistant Professor, Department of Electronics and Telecommunication,

More information

The Insight Toolkit. Image Registration Algorithms & Frameworks

The Insight Toolkit. Image Registration Algorithms & Frameworks The Insight Toolkit Image Registration Algorithms & Frameworks Registration in ITK Image Registration Framework Multi Resolution Registration Framework Components PDE Based Registration FEM Based Registration

More information

Basic principles of MR image analysis. Basic principles of MR image analysis. Basic principles of MR image analysis

Basic principles of MR image analysis. Basic principles of MR image analysis. Basic principles of MR image analysis Basic principles of MR image analysis Basic principles of MR image analysis Julien Milles Leiden University Medical Center Terminology of fmri Brain extraction Registration Linear registration Non-linear

More information

DRAMMS: Deformable Registration via Attribute Matching and Mutual-Saliency weighting

DRAMMS: Deformable Registration via Attribute Matching and Mutual-Saliency weighting DRAMMS: Deformable Registration via Attribute Matching and Mutual-Saliency weighting Yangming Ou, Christos Davatzikos Section of Biomedical Image Analysis (SBIA) University of Pennsylvania Outline 1. Background

More information

Correspondence Detection Using Wavelet-Based Attribute Vectors

Correspondence Detection Using Wavelet-Based Attribute Vectors Correspondence Detection Using Wavelet-Based Attribute Vectors Zhong Xue, Dinggang Shen, and Christos Davatzikos Section of Biomedical Image Analysis, Department of Radiology University of Pennsylvania,

More information

Digital Volume Correlation for Materials Characterization

Digital Volume Correlation for Materials Characterization 19 th World Conference on Non-Destructive Testing 2016 Digital Volume Correlation for Materials Characterization Enrico QUINTANA, Phillip REU, Edward JIMENEZ, Kyle THOMPSON, Sharlotte KRAMER Sandia National

More information

MIXTURE MODELING FOR DIGITAL MAMMOGRAM DISPLAY AND ANALYSIS 1

MIXTURE MODELING FOR DIGITAL MAMMOGRAM DISPLAY AND ANALYSIS 1 MIXTURE MODELING FOR DIGITAL MAMMOGRAM DISPLAY AND ANALYSIS 1 Stephen R. Aylward, Bradley M. Hemminger, and Etta D. Pisano Department of Radiology Medical Image Display and Analysis Group University of

More information

Predicting Tumour Location by Modelling the Deformation of the Breast using Nonlinear Elasticity

Predicting Tumour Location by Modelling the Deformation of the Breast using Nonlinear Elasticity Predicting Tumour Location by Modelling the Deformation of the Breast using Nonlinear Elasticity November 8th, 2006 Outline Motivation Motivation Motivation for Modelling Breast Deformation Mesh Generation

More information

Non-rigid Image Registration

Non-rigid Image Registration Overview Non-rigid Image Registration Introduction to image registration - he goal of image registration - Motivation for medical image registration - Classification of image registration - Nonrigid registration

More information

Foveal Algorithm for the Detection of Microcalcification Clusters: a FROC Analysis

Foveal Algorithm for the Detection of Microcalcification Clusters: a FROC Analysis Foveal Algorithm for the Detection of Microcalcification Clusters: a FROC Analysis Marius George Linguraru a,b, Michael Brady a and Ruth English c a Medical Vision Laboratory, University of Oxford, Ewert

More information

Nonrigid Motion Compensation of Free Breathing Acquired Myocardial Perfusion Data

Nonrigid Motion Compensation of Free Breathing Acquired Myocardial Perfusion Data Nonrigid Motion Compensation of Free Breathing Acquired Myocardial Perfusion Data Gert Wollny 1, Peter Kellman 2, Andrés Santos 1,3, María-Jesus Ledesma 1,3 1 Biomedical Imaging Technologies, Department

More information

Bildverarbeitung für die Medizin 2007

Bildverarbeitung für die Medizin 2007 Bildverarbeitung für die Medizin 2007 Image Registration with Local Rigidity Constraints Jan Modersitzki Institute of Mathematics, University of Lübeck, Wallstraße 40, D-23560 Lübeck 1 Summary Registration

More information

Multi-Atlas Brain MRI Segmentation with Multiway Cut

Multi-Atlas Brain MRI Segmentation with Multiway Cut Multi-Atlas Brain MRI Segmentation with Multiway Cut Duygu Sarikaya, Liang Zhao, and Jason J. Corso SUNY Buffalo, Computer Science and Engineering Department, 338 Davis Hall- Buffalo, New York, USA 14260-2500

More information

Robust Realignment of fmri Time Series Data

Robust Realignment of fmri Time Series Data Robust Realignment of fmri Time Series Data Ben Dodson bjdodson@stanford.edu Olafur Gudmundsson olafurg@stanford.edu December 12, 2008 Abstract FMRI data has become an increasingly popular source for exploring

More information

An ITK Filter for Bayesian Segmentation: itkbayesianclassifierimagefilter

An ITK Filter for Bayesian Segmentation: itkbayesianclassifierimagefilter An ITK Filter for Bayesian Segmentation: itkbayesianclassifierimagefilter John Melonakos 1, Karthik Krishnan 2 and Allen Tannenbaum 1 1 Georgia Institute of Technology, Atlanta GA 30332, USA {jmelonak,

More information

K-Means Clustering Using Localized Histogram Analysis

K-Means Clustering Using Localized Histogram Analysis K-Means Clustering Using Localized Histogram Analysis Michael Bryson University of South Carolina, Department of Computer Science Columbia, SC brysonm@cse.sc.edu Abstract. The first step required for many

More information

Non-rigid Image Registration using Electric Current Flow

Non-rigid Image Registration using Electric Current Flow Non-rigid Image Registration using Electric Current Flow Shu Liao, Max W. K. Law and Albert C. S. Chung Lo Kwee-Seong Medical Image Analysis Laboratory, Department of Computer Science and Engineering,

More information

Automatic Generation of Training Data for Brain Tissue Classification from MRI

Automatic Generation of Training Data for Brain Tissue Classification from MRI MICCAI-2002 1 Automatic Generation of Training Data for Brain Tissue Classification from MRI Chris A. Cocosco, Alex P. Zijdenbos, and Alan C. Evans McConnell Brain Imaging Centre, Montreal Neurological

More information

Image Segmentation and Registration

Image Segmentation and Registration Image Segmentation and Registration Dr. Christine Tanner (tanner@vision.ee.ethz.ch) Computer Vision Laboratory, ETH Zürich Dr. Verena Kaynig, Machine Learning Laboratory, ETH Zürich Outline Segmentation

More information

Spatio-Temporal Registration of Biomedical Images by Computational Methods

Spatio-Temporal Registration of Biomedical Images by Computational Methods Spatio-Temporal Registration of Biomedical Images by Computational Methods Francisco P. M. Oliveira, João Manuel R. S. Tavares tavares@fe.up.pt, www.fe.up.pt/~tavares Outline 1. Introduction 2. Spatial

More information

Adaptive Local Multi-Atlas Segmentation: Application to Heart Segmentation in Chest CT Scans

Adaptive Local Multi-Atlas Segmentation: Application to Heart Segmentation in Chest CT Scans Adaptive Local Multi-Atlas Segmentation: Application to Heart Segmentation in Chest CT Scans Eva M. van Rikxoort, Ivana Isgum, Marius Staring, Stefan Klein and Bram van Ginneken Image Sciences Institute,

More information

Registration by continuous optimisation. Stefan Klein Erasmus MC, the Netherlands Biomedical Imaging Group Rotterdam (BIGR)

Registration by continuous optimisation. Stefan Klein Erasmus MC, the Netherlands Biomedical Imaging Group Rotterdam (BIGR) Registration by continuous optimisation Stefan Klein Erasmus MC, the Netherlands Biomedical Imaging Group Rotterdam (BIGR) Registration = optimisation C t x t y 1 Registration = optimisation C t x t y

More information

Positioning Lesion from Breast MRI and Mammogram using Registration Method

Positioning Lesion from Breast MRI and Mammogram using Registration Method Positioning Lesion from Breast MRI and Mammogram using Registration Method Rugma M. *, Malu G., C. Gopakumar, Elizabeth Sherly * PG Scholar, Dept. of CE, College of Engineering, Chengannur. Alapuzha, India

More information

Lesion Segmentation and Bias Correction in Breast Ultrasound B-mode Images Including Elastography Information

Lesion Segmentation and Bias Correction in Breast Ultrasound B-mode Images Including Elastography Information Lesion Segmentation and Bias Correction in Breast Ultrasound B-mode Images Including Elastography Information Gerard Pons a, Joan Martí a, Robert Martí a, Mariano Cabezas a, Andrew di Battista b, and J.

More information

Classification of Subject Motion for Improved Reconstruction of Dynamic Magnetic Resonance Imaging

Classification of Subject Motion for Improved Reconstruction of Dynamic Magnetic Resonance Imaging 1 CS 9 Final Project Classification of Subject Motion for Improved Reconstruction of Dynamic Magnetic Resonance Imaging Feiyu Chen Department of Electrical Engineering ABSTRACT Subject motion is a significant

More information

MULTIVIEW REPRESENTATION OF 3D OBJECTS OF A SCENE USING VIDEO SEQUENCES

MULTIVIEW REPRESENTATION OF 3D OBJECTS OF A SCENE USING VIDEO SEQUENCES MULTIVIEW REPRESENTATION OF 3D OBJECTS OF A SCENE USING VIDEO SEQUENCES Mehran Yazdi and André Zaccarin CVSL, Dept. of Electrical and Computer Engineering, Laval University Ste-Foy, Québec GK 7P4, Canada

More information

Deformable Registration Using Scale Space Keypoints

Deformable Registration Using Scale Space Keypoints Deformable Registration Using Scale Space Keypoints Mehdi Moradi a, Purang Abolmaesoumi a,b and Parvin Mousavi a a School of Computing, Queen s University, Kingston, Ontario, Canada K7L 3N6; b Department

More information

A Functional Connectivity Inspired Approach to Non-Local fmri Analysis

A Functional Connectivity Inspired Approach to Non-Local fmri Analysis A Functional Connectivity Inspired Approach to Non-Local fmri Analysis Anders Eklund, Mats Andersson and Hans Knutsson Linköping University Post Print N.B.: When citing this work, cite the original article.

More information

Towards Breast Anatomy Simulation Using GPUs

Towards Breast Anatomy Simulation Using GPUs Towards Breast Anatomy Simulation Using GPUs Joseph H. Chui 1, David D. Pokrajac 2, Andrew D.A. Maidment 3, and Predrag R. Bakic 4 1 Department of Radiology, University of Pennsylvania, Philadelphia PA

More information

Multi-atlas labeling with population-specific template and non-local patch-based label fusion

Multi-atlas labeling with population-specific template and non-local patch-based label fusion Multi-atlas labeling with population-specific template and non-local patch-based label fusion Vladimir Fonov, Pierrick Coupé, Simon Eskildsen, Jose Manjon, Louis Collins To cite this version: Vladimir

More information

Predicting Tumour Location by Simulating Large Deformations of the Breast using a 3D Finite Element Model and Nonlinear Elasticity

Predicting Tumour Location by Simulating Large Deformations of the Breast using a 3D Finite Element Model and Nonlinear Elasticity Predicting Tumour Location by Simulating Large Deformations of the Breast using a 3D Finite Element Model and Nonlinear Elasticity P. Pathmanathan 1 D. Gavaghan 1 J. Whiteley 1 M. Brady 2 M. Nash 3 P.

More information

ELEC Dr Reji Mathew Electrical Engineering UNSW

ELEC Dr Reji Mathew Electrical Engineering UNSW ELEC 4622 Dr Reji Mathew Electrical Engineering UNSW Review of Motion Modelling and Estimation Introduction to Motion Modelling & Estimation Forward Motion Backward Motion Block Motion Estimation Motion

More information

Scene-Based Segmentation of Multiple Muscles from MRI in MITK

Scene-Based Segmentation of Multiple Muscles from MRI in MITK Scene-Based Segmentation of Multiple Muscles from MRI in MITK Yan Geng 1, Sebastian Ullrich 2, Oliver Grottke 3, Rolf Rossaint 3, Torsten Kuhlen 2, Thomas M. Deserno 1 1 Department of Medical Informatics,

More information

An Experimental Investigation into the Rank Function of the Heterogeneous Earliest Finish Time Scheduling Algorithm

An Experimental Investigation into the Rank Function of the Heterogeneous Earliest Finish Time Scheduling Algorithm An Experimental Investigation into the Rank Function of the Heterogeneous Earliest Finish Time Scheduling Algorithm Henan Zhao and Rizos Sakellariou Department of Computer Science, University of Manchester,

More information

ADAPTIVE GRAPH CUTS WITH TISSUE PRIORS FOR BRAIN MRI SEGMENTATION

ADAPTIVE GRAPH CUTS WITH TISSUE PRIORS FOR BRAIN MRI SEGMENTATION ADAPTIVE GRAPH CUTS WITH TISSUE PRIORS FOR BRAIN MRI SEGMENTATION Abstract: MIP Project Report Spring 2013 Gaurav Mittal 201232644 This is a detailed report about the course project, which was to implement

More information

Methodological progress in image registration for ventilation estimation, segmentation propagation and multi-modal fusion

Methodological progress in image registration for ventilation estimation, segmentation propagation and multi-modal fusion Methodological progress in image registration for ventilation estimation, segmentation propagation and multi-modal fusion Mattias P. Heinrich Julia A. Schnabel, Mark Jenkinson, Sir Michael Brady 2 Clinical

More information

Non-Rigid Multimodal Medical Image Registration using Optical Flow and Gradient Orientation

Non-Rigid Multimodal Medical Image Registration using Optical Flow and Gradient Orientation M. HEINRICH et al.: MULTIMODAL REGISTRATION USING GRADIENT ORIENTATION 1 Non-Rigid Multimodal Medical Image Registration using Optical Flow and Gradient Orientation Mattias P. Heinrich 1 mattias.heinrich@eng.ox.ac.uk

More information

MARS: Multiple Atlases Robust Segmentation

MARS: Multiple Atlases Robust Segmentation Software Release (1.0.1) Last updated: April 30, 2014. MARS: Multiple Atlases Robust Segmentation Guorong Wu, Minjeong Kim, Gerard Sanroma, and Dinggang Shen {grwu, mjkim, gerard_sanroma, dgshen}@med.unc.edu

More information

A Study of Medical Image Analysis System

A Study of Medical Image Analysis System Indian Journal of Science and Technology, Vol 8(25), DOI: 10.17485/ijst/2015/v8i25/80492, October 2015 ISSN (Print) : 0974-6846 ISSN (Online) : 0974-5645 A Study of Medical Image Analysis System Kim Tae-Eun

More information

Pathology Hinting as the Combination of Automatic Segmentation with a Statistical Shape Model

Pathology Hinting as the Combination of Automatic Segmentation with a Statistical Shape Model Pathology Hinting as the Combination of Automatic Segmentation with a Statistical Shape Model Pascal A. Dufour 1,2, Hannan Abdillahi 3, Lala Ceklic 3, Ute Wolf-Schnurrbusch 2,3, and Jens Kowal 1,2 1 ARTORG

More information

SPM8 for Basic and Clinical Investigators. Preprocessing. fmri Preprocessing

SPM8 for Basic and Clinical Investigators. Preprocessing. fmri Preprocessing SPM8 for Basic and Clinical Investigators Preprocessing fmri Preprocessing Slice timing correction Geometric distortion correction Head motion correction Temporal filtering Intensity normalization Spatial

More information

Introduction to Medical Image Processing

Introduction to Medical Image Processing Introduction to Medical Image Processing Δ Essential environments of a medical imaging system Subject Image Analysis Energy Imaging System Images Image Processing Feature Images Image processing may be

More information

2 Michael E. Leventon and Sarah F. F. Gibson a b c d Fig. 1. (a, b) Two MR scans of a person's knee. Both images have high resolution in-plane, but ha

2 Michael E. Leventon and Sarah F. F. Gibson a b c d Fig. 1. (a, b) Two MR scans of a person's knee. Both images have high resolution in-plane, but ha Model Generation from Multiple Volumes using Constrained Elastic SurfaceNets Michael E. Leventon and Sarah F. F. Gibson 1 MIT Artificial Intelligence Laboratory, Cambridge, MA 02139, USA leventon@ai.mit.edu

More information

Hybrid Spline-based Multimodal Registration using a Local Measure for Mutual Information

Hybrid Spline-based Multimodal Registration using a Local Measure for Mutual Information Hybrid Spline-based Multimodal Registration using a Local Measure for Mutual Information Andreas Biesdorf 1, Stefan Wörz 1, Hans-Jürgen Kaiser 2, Karl Rohr 1 1 University of Heidelberg, BIOQUANT, IPMB,

More information

3D Tongue Motion from Tagged and Cine MR Images

3D Tongue Motion from Tagged and Cine MR Images 3D Tongue Motion from Tagged and Cine MR Images Fangxu Xing 1, Jonghye Woo 1,2,EmiZ.Murano 3, Junghoon Lee 1,4, Maureen Stone 2, and Jerry L. Prince 1 1 Department of Electrical and Computer Engineering,

More information

Fast 3D Mean Shift Filter for CT Images

Fast 3D Mean Shift Filter for CT Images Fast 3D Mean Shift Filter for CT Images Gustavo Fernández Domínguez, Horst Bischof, and Reinhard Beichel Institute for Computer Graphics and Vision, Graz University of Technology Inffeldgasse 16/2, A-8010,

More information

Preprocessing II: Between Subjects John Ashburner

Preprocessing II: Between Subjects John Ashburner Preprocessing II: Between Subjects John Ashburner Pre-processing Overview Statistics or whatever fmri time-series Anatomical MRI Template Smoothed Estimate Spatial Norm Motion Correct Smooth Coregister

More information

Segmentation and Classification of Breast Tumor Using Dynamic Contrast-Enhanced MR Images

Segmentation and Classification of Breast Tumor Using Dynamic Contrast-Enhanced MR Images Segmentation and Classification of Breast Tumor Using Dynamic Contrast-Enhanced MR Images Yuanjie Zheng, Sajjad Baloch, Sarah Englander, Mitchell D. Schnall, and Dinggang Shen Department of Radiology,

More information

Segmentation of the pectoral muscle edge on mammograms by tunable parametric edge detection

Segmentation of the pectoral muscle edge on mammograms by tunable parametric edge detection Segmentation of the pectoral muscle edge on mammograms by tunable parametric edge detection R CHANDRASEKHAR and Y ATTIKIOUZEL Australian Research Centre for Medical Engineering (ARCME) The University of

More information

Basic fmri Design and Analysis. Preprocessing

Basic fmri Design and Analysis. Preprocessing Basic fmri Design and Analysis Preprocessing fmri Preprocessing Slice timing correction Geometric distortion correction Head motion correction Temporal filtering Intensity normalization Spatial filtering

More information

Intraoperative Prostate Tracking with Slice-to-Volume Registration in MR

Intraoperative Prostate Tracking with Slice-to-Volume Registration in MR Intraoperative Prostate Tracking with Slice-to-Volume Registration in MR Sean Gill a, Purang Abolmaesumi a,b, Siddharth Vikal a, Parvin Mousavi a and Gabor Fichtinger a,b,* (a) School of Computing, Queen

More information

Using Probability Maps for Multi organ Automatic Segmentation

Using Probability Maps for Multi organ Automatic Segmentation Using Probability Maps for Multi organ Automatic Segmentation Ranveer Joyseeree 1,2, Óscar Jiménez del Toro1, and Henning Müller 1,3 1 University of Applied Sciences Western Switzerland (HES SO), Sierre,

More information

Mass Classification Method in Mammogram Using Fuzzy K-Nearest Neighbour Equality

Mass Classification Method in Mammogram Using Fuzzy K-Nearest Neighbour Equality Mass Classification Method in Mammogram Using Fuzzy K-Nearest Neighbour Equality Abstract: Mass classification of objects is an important area of research and application in a variety of fields. In this

More information

BLUT : Fast and Low Memory B-spline Image Interpolation

BLUT : Fast and Low Memory B-spline Image Interpolation BLUT : Fast and Low Memory B-spline Image Interpolation David Sarrut a,b,c, Jef Vandemeulebroucke a,b,c a Université de Lyon, F-69622 Lyon, France. b Creatis, CNRS UMR 5220, F-69622, Villeurbanne, France.

More information

Multi-Atlas Segmentation of the Cardiac MR Right Ventricle

Multi-Atlas Segmentation of the Cardiac MR Right Ventricle Multi-Atlas Segmentation of the Cardiac MR Right Ventricle Yangming Ou, Jimit Doshi, Guray Erus, and Christos Davatzikos Section of Biomedical Image Analysis (SBIA) Department of Radiology, University

More information

A ranklet-based CAD for digital mammography

A ranklet-based CAD for digital mammography A ranklet-based CAD for digital mammography Enrico Angelini 1, Renato Campanini 1, Emiro Iampieri 1, Nico Lanconelli 1, Matteo Masotti 1, Todor Petkov 1, and Matteo Roffilli 2 1 Physics Department, University

More information

Neuroimaging and mathematical modelling Lesson 2: Voxel Based Morphometry

Neuroimaging and mathematical modelling Lesson 2: Voxel Based Morphometry Neuroimaging and mathematical modelling Lesson 2: Voxel Based Morphometry Nivedita Agarwal, MD Nivedita.agarwal@apss.tn.it Nivedita.agarwal@unitn.it Volume and surface morphometry Brain volume White matter

More information

Medical images, segmentation and analysis

Medical images, segmentation and analysis Medical images, segmentation and analysis ImageLab group http://imagelab.ing.unimo.it Università degli Studi di Modena e Reggio Emilia Medical Images Macroscopic Dermoscopic ELM enhance the features of

More information

Functional MRI in Clinical Research and Practice Preprocessing

Functional MRI in Clinical Research and Practice Preprocessing Functional MRI in Clinical Research and Practice Preprocessing fmri Preprocessing Slice timing correction Geometric distortion correction Head motion correction Temporal filtering Intensity normalization

More information

Image Registration. Prof. Dr. Lucas Ferrari de Oliveira UFPR Informatics Department

Image Registration. Prof. Dr. Lucas Ferrari de Oliveira UFPR Informatics Department Image Registration Prof. Dr. Lucas Ferrari de Oliveira UFPR Informatics Department Introduction Visualize objects inside the human body Advances in CS methods to diagnosis, treatment planning and medical

More information

EPI Data Are Acquired Serially. EPI Data Are Acquired Serially 10/23/2011. Functional Connectivity Preprocessing. fmri Preprocessing

EPI Data Are Acquired Serially. EPI Data Are Acquired Serially 10/23/2011. Functional Connectivity Preprocessing. fmri Preprocessing Functional Connectivity Preprocessing Geometric distortion Head motion Geometric distortion Head motion EPI Data Are Acquired Serially EPI Data Are Acquired Serially descending 1 EPI Data Are Acquired

More information

Mobile Human Detection Systems based on Sliding Windows Approach-A Review

Mobile Human Detection Systems based on Sliding Windows Approach-A Review Mobile Human Detection Systems based on Sliding Windows Approach-A Review Seminar: Mobile Human detection systems Njieutcheu Tassi cedrique Rovile Department of Computer Engineering University of Heidelberg

More information

Shape-based Diffeomorphic Registration on Hippocampal Surfaces Using Beltrami Holomorphic Flow

Shape-based Diffeomorphic Registration on Hippocampal Surfaces Using Beltrami Holomorphic Flow Shape-based Diffeomorphic Registration on Hippocampal Surfaces Using Beltrami Holomorphic Flow Abstract. Finding meaningful 1-1 correspondences between hippocampal (HP) surfaces is an important but difficult

More information

CHAPTER 2. Morphometry on rodent brains. A.E.H. Scheenstra J. Dijkstra L. van der Weerd

CHAPTER 2. Morphometry on rodent brains. A.E.H. Scheenstra J. Dijkstra L. van der Weerd CHAPTER 2 Morphometry on rodent brains A.E.H. Scheenstra J. Dijkstra L. van der Weerd This chapter was adapted from: Volumetry and other quantitative measurements to assess the rodent brain, In vivo NMR

More information

Fully Automatic Multi-organ Segmentation based on Multi-boost Learning and Statistical Shape Model Search

Fully Automatic Multi-organ Segmentation based on Multi-boost Learning and Statistical Shape Model Search Fully Automatic Multi-organ Segmentation based on Multi-boost Learning and Statistical Shape Model Search Baochun He, Cheng Huang, Fucang Jia Shenzhen Institutes of Advanced Technology, Chinese Academy

More information

Integration of Valley Orientation Distribution for Polyp Region Identification in Colonoscopy

Integration of Valley Orientation Distribution for Polyp Region Identification in Colonoscopy Integration of Valley Orientation Distribution for Polyp Region Identification in Colonoscopy Jorge Bernal, Javier Sánchez, and Fernando Vilariño Computer Vision Centre and Computer Science Department,

More information

Medical Image Segmentation Based on Mutual Information Maximization

Medical Image Segmentation Based on Mutual Information Maximization Medical Image Segmentation Based on Mutual Information Maximization J.Rigau, M.Feixas, M.Sbert, A.Bardera, and I.Boada Institut d Informatica i Aplicacions, Universitat de Girona, Spain {jaume.rigau,miquel.feixas,mateu.sbert,anton.bardera,imma.boada}@udg.es

More information

White Matter Lesion Segmentation (WMLS) Manual

White Matter Lesion Segmentation (WMLS) Manual White Matter Lesion Segmentation (WMLS) Manual 1. Introduction White matter lesions (WMLs) are brain abnormalities that appear in different brain diseases, such as multiple sclerosis (MS), head injury,

More information

Dealing with Inaccurate Face Detection for Automatic Gender Recognition with Partially Occluded Faces

Dealing with Inaccurate Face Detection for Automatic Gender Recognition with Partially Occluded Faces Dealing with Inaccurate Face Detection for Automatic Gender Recognition with Partially Occluded Faces Yasmina Andreu, Pedro García-Sevilla, and Ramón A. Mollineda Dpto. Lenguajes y Sistemas Informáticos

More information

Model-Based Human Motion Capture from Monocular Video Sequences

Model-Based Human Motion Capture from Monocular Video Sequences Model-Based Human Motion Capture from Monocular Video Sequences Jihun Park 1, Sangho Park 2, and J.K. Aggarwal 2 1 Department of Computer Engineering Hongik University Seoul, Korea jhpark@hongik.ac.kr

More information

Non-Parametric Bayesian Registration (NParBR) on CT Lungs Data - EMPIRE10 Challenge

Non-Parametric Bayesian Registration (NParBR) on CT Lungs Data - EMPIRE10 Challenge Non-Parametric Bayesian Registration (NParBR) on CT Lungs Data - EMPIRE10 Challenge David Pilutti 1, Maddalena Strumia 1, Stathis Hadjidemetriou 2 1 University Medical Center Freiburg, 79106 Freiburg,

More information

Knowledge-Based Segmentation of Brain MRI Scans Using the Insight Toolkit

Knowledge-Based Segmentation of Brain MRI Scans Using the Insight Toolkit Knowledge-Based Segmentation of Brain MRI Scans Using the Insight Toolkit John Melonakos 1, Ramsey Al-Hakim 1, James Fallon 2 and Allen Tannenbaum 1 1 Georgia Institute of Technology, Atlanta GA 30332,

More information

Non-Rigid Image Registration III

Non-Rigid Image Registration III Non-Rigid Image Registration III CS6240 Multimedia Analysis Leow Wee Kheng Department of Computer Science School of Computing National University of Singapore Leow Wee Kheng (CS6240) Non-Rigid Image Registration

More information

Chapter 9 Object Tracking an Overview

Chapter 9 Object Tracking an Overview Chapter 9 Object Tracking an Overview The output of the background subtraction algorithm, described in the previous chapter, is a classification (segmentation) of pixels into foreground pixels (those belonging

More information

WorkstationOne. - a diagnostic breast imaging workstation. User Training Presentation. doc #24 (v2.0) Let MammoOne Assist You in Digital Mammography

WorkstationOne. - a diagnostic breast imaging workstation. User Training Presentation. doc #24 (v2.0) Let MammoOne Assist You in Digital Mammography WorkstationOne - a diagnostic breast imaging workstation User Training Presentation WorkstationOne Unique streamlined workflow for efficient digital mammography reading Let MammoOne Let MammoOne Assist

More information

Representation of 2D objects with a topology preserving network

Representation of 2D objects with a topology preserving network Representation of 2D objects with a topology preserving network Francisco Flórez, Juan Manuel García, José García, Antonio Hernández, Departamento de Tecnología Informática y Computación. Universidad de

More information

A Novel Criterion Function in Feature Evaluation. Application to the Classification of Corks.

A Novel Criterion Function in Feature Evaluation. Application to the Classification of Corks. A Novel Criterion Function in Feature Evaluation. Application to the Classification of Corks. X. Lladó, J. Martí, J. Freixenet, Ll. Pacheco Computer Vision and Robotics Group Institute of Informatics and

More information

Graph Matching Iris Image Blocks with Local Binary Pattern

Graph Matching Iris Image Blocks with Local Binary Pattern Graph Matching Iris Image Blocs with Local Binary Pattern Zhenan Sun, Tieniu Tan, and Xianchao Qiu Center for Biometrics and Security Research, National Laboratory of Pattern Recognition, Institute of

More information

Differential Structure in non-linear Image Embedding Functions

Differential Structure in non-linear Image Embedding Functions Differential Structure in non-linear Image Embedding Functions Robert Pless Department of Computer Science, Washington University in St. Louis pless@cse.wustl.edu Abstract Many natural image sets are samples

More information

Combinatorial optimization and its applications in image Processing. Filip Malmberg

Combinatorial optimization and its applications in image Processing. Filip Malmberg Combinatorial optimization and its applications in image Processing Filip Malmberg Part 1: Optimization in image processing Optimization in image processing Many image processing problems can be formulated

More information

Introduction to Medical Image Registration

Introduction to Medical Image Registration Introduction to Medical Image Registration Sailesh Conjeti Computer Aided Medical Procedures (CAMP), Technische Universität München, Germany sailesh.conjeti@tum.de Partially adapted from slides by: 1.

More information

3D VISUALIZATION OF SEGMENTED CRUCIATE LIGAMENTS 1. INTRODUCTION

3D VISUALIZATION OF SEGMENTED CRUCIATE LIGAMENTS 1. INTRODUCTION JOURNAL OF MEDICAL INFORMATICS & TECHNOLOGIES Vol. 10/006, ISSN 164-6037 Paweł BADURA * cruciate ligament, segmentation, fuzzy connectedness,3d visualization 3D VISUALIZATION OF SEGMENTED CRUCIATE LIGAMENTS

More information

Edge-Preserving MRI Super Resolution Using a High Frequency Regularization Technique

Edge-Preserving MRI Super Resolution Using a High Frequency Regularization Technique Edge-Preserving MRI Super Resolution Using a High Frequency Regularization Technique Kaveh Ahmadi Department of EECS University of Toledo, Toledo, Ohio, USA 43606 Email: Kaveh.ahmadi@utoledo.edu Ezzatollah

More information

Auto-contouring the Prostate for Online Adaptive Radiotherapy

Auto-contouring the Prostate for Online Adaptive Radiotherapy Auto-contouring the Prostate for Online Adaptive Radiotherapy Yan Zhou 1 and Xiao Han 1 Elekta Inc., Maryland Heights, MO, USA yan.zhou@elekta.com, xiao.han@elekta.com, Abstract. Among all the organs under

More information

Simultaneous surface texture classification and illumination tilt angle prediction

Simultaneous surface texture classification and illumination tilt angle prediction Simultaneous surface texture classification and illumination tilt angle prediction X. Lladó, A. Oliver, M. Petrou, J. Freixenet, and J. Martí Computer Vision and Robotics Group - IIiA. University of Girona

More information

HST.582J / 6.555J / J Biomedical Signal and Image Processing Spring 2007

HST.582J / 6.555J / J Biomedical Signal and Image Processing Spring 2007 MIT OpenCourseWare http://ocw.mit.edu HST.582J / 6.555J / 16.456J Biomedical Signal and Image Processing Spring 2007 For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.

More information