Fereshteh S. Bashiri Advisors: Zeyun Yu, Roshan M. D’souza

Slides:



Advertisements
Similar presentations
Applications of one-class classification
Advertisements

Image Registration  Mapping of Evolution. Registration Goals Assume the correspondences are known Find such f() and g() such that the images are best.
Mutual Information Based Registration of Medical Images
The content of these slides by John Galeotti, © Carnegie Mellon University (CMU), was made possible in part by NIH NLM contract# HHSN P,
Treatment Planning of HIFU: Rigid Registration of MRI to Ultrasound Kidney Images Tara Yates 1, Penny Probert Smith 1, J. Alison Noble 1, Tom Leslie 2,
Coregistration and Normalisation By Lieke de Boer & Julie Guerin.
Automatic Feature Extraction for Multi-view 3D Face Recognition
A Similarity Retrieval System for Multimodal Functional Brain Images Rosalia F. Tungaraza Advisor: Prof. Linda G. Shapiro Ph.D. Defense Computer Science.
A Computationally Efficient Approach for 2D-3D Image Registration Juri Minxha Medical Image Analysis Professor Benjamin Kimia Spring 2011 Brown University.
A Computationally Efficient Approach for 2D-3D Image Registration Juri Minxha Medical Image Analysis Professor Benjamin Kimia Spring 2011 Brown University.
1 Improving Entropy Registration Theodor D. Richardson.
Incremental Learning of Temporally-Coherent Gaussian Mixture Models Ognjen Arandjelović, Roberto Cipolla Engineering Department, University of Cambridge.
Mutual Information for Image Registration and Feature Selection
Yujun Guo Kent State University August PRESENTATION A Binarization Approach for CT-MR Registration Using Normalized Mutual Information.
Image Registration Narendhran Vijayakumar (Naren) 12/17/2007 Department of Electrical and Computer Engineering 1.
Medical Image Registration
Step 1: Load data we have 20 axial slices for the brain starting from the neck up to top of the head for each of three modalities: T1, T1C, T2.
Mutual Information Narendhran Vijayakumar 03/14/2008.
Lijuan Zhao Advisors: Prof. Fatima Merchant Prof. Shishir Shah.
My Research Experience Cheng Qian. Outline 3D Reconstruction Based on Range Images Color Engineering Thermal Image Restoration.
Manifold learning: Locally Linear Embedding Jieping Ye Department of Computer Science and Engineering Arizona State University
Mutual Information-based Stereo Matching Combined with SIFT Descriptor in Log-chromaticity Color Space Yong Seok Heo, Kyoung Mu Lee, and Sang Uk Lee.
Medical Image Analysis Image Enhancement Figures come from the textbook: Medical Image Analysis, by Atam P. Dhawan, IEEE Press, 2003.
The Bernard M. Gordon Center for Subsurface Sensing and Imaging Systems Charles V. Stewart Dept. of Computer Science Rensselaer Poly. Inst. CenSSIS Charles.
Medical Image Analysis Image Reconstruction Figures come from the textbook: Medical Image Analysis, by Atam P. Dhawan, IEEE Press, 2003.
CSci 6971: Image Registration Lecture 3: Images and Transformations March 1, 2005 Prof. Charlene Tsai.
Jan Kamenický Mariánská  We deal with medical images ◦ Different viewpoints - multiview ◦ Different times - multitemporal ◦ Different sensors.
MEDICAL IMAGE REGISTRATION BY MAXIMIZATION OF MUTUAL INFORMATION Dissertation Defense by Chi-hsiang Lo June 27, 2003 PRESENTATION.
Medical Image Analysis Image Registration Figures come from the textbook: Medical Image Analysis, by Atam P. Dhawan, IEEE Press, 2003.
Spatio-Temporal Free-Form Registration of Cardiac MR Image Sequences Antonios Perperidis s /02/2006.
A New Method of Probability Density Estimation for Mutual Information Based Image Registration Ajit Rajwade, Arunava Banerjee, Anand Rangarajan. Dept.
Conclusions The success rate of proposed method is higher than that of traditional MI MI based on GVFI is robust to noise GVFI based on f1 performs better.
M. Pokric, P.A. Bromiley, N.A. Thacker, M.L.J. Scott, and A. Jackson University of Manchester Imaging Science and Biomedical Engineering Probabilistic.
AdvisorStudent Dr. Jia Li Shaojun Liu Dept. of Computer Science and Engineering, Oakland University Automatic 3D Image Segmentation of Internal Lung Structures.
A B C D E F A ABSTRACT A novel, efficient, robust, feature-based algorithm is presented for intramodality and multimodality medical image registration.
Statistical Parametric Mapping Lecture 11 - Chapter 13 Head motion and correction Textbook: Functional MRI an introduction to methods, Peter Jezzard, Paul.
Group-wise Registration in NAMIC-kit Serdar K Balci (MIT) Lilla Zöllei (MGH) Kinh Tieu (BWH) Mert R Sabuncu (MIT) Polina Golland (MIT)
MultiModality Registration Using Hilbert-Schmidt Estimators By: Srinivas Peddi Computer Integrated Surgery II April 6 th, 2001.
National Alliance for Medical Image Computing Hierarchical Atlas Based EM Segmentation.
Photoconsistency constraint C2 q C1 p l = 2 l = 3 Depth labels If this 3D point is visible in both cameras, pixels p and q should have similar intensities.
A 2D/3D correspondence building method for reconstruction of a 3D bone surface model Longwei Fang
Date of download: 7/9/2016 Copyright © 2016 SPIE. All rights reserved. An example of fMRI data acquired from a healthy fetus in the coronal plane. It includes.
A Bayesian Approach for Transformation Estimation Camille Izard and Bruno Jedynak Landmark Detection in brain MRI Laboratoire Paul Painlevé Université.
Yun, Hyuk Jin. Theory A.Nonuniformity Model where at location x, v is the measured signal, u is the true signal emitted by the tissue, is an unknown.
EXAMPLE 3-1 Estimating the Cost of a College Degree
Segmentation of Single-Figure Objects by Deformable M-reps
Medical Image Analysis
Statistical-Mechanical Approach to Probabilistic Image Processing -- Loopy Belief Propagation and Advanced Mean-Field Method -- Kazuyuki Tanaka and Noriko.
Shuang Hong Yang College of Computing, Georgia Tech, USA Hongyuan Zha
University of Ioannina
Semi-Global Matching with self-adjusting penalties
Mutual Information Based Registration of Medical Images
Presenter: Hajar Emami
Multi-modality image registration using mutual information based on gradient vector flow Yujun Guo May 1,2006.
Outline Nonlinear Dimension Reduction Brief introduction Isomap LLE
Computational Neuroanatomy for Dummies
Corresponding midsagittal CT (left), MR (middle), and registered (right) images of the cervical spine show proper alignment and the relationship between.
Graph Theoretic Analysis of Resting State Functional MR Imaging
From buttons to code Eamonn Walsh & Domenica Bueti
Image Registration 박성진.
Nonparametric Hypothesis Tests for Dependency Structures
Anatomical Measures John Ashburner
D.D. Anderson, A.M. Kern, T.J. Stockman, C.M. Findlay, N.A. Segal 
Xiaomo Chen, Marc Zirnsak, Tirin Moore  Cell Reports 
Sam C. Berens, Jessica S. Horst, Chris M. Bird  Current Biology 
MultiModality Registration using Hilbert-Schmidt Estimators
--- Range Image Registration
Image Registration  Mapping of Evolution
Registration Foundations
Strength of relation High Low Number of data Relationship Data
Presentation transcript:

Partial to Full multi-modal medical image registration based on structural representation Fereshteh S. Bashiri Advisors: Zeyun Yu, Roshan M. D’souza College of Engineering and Applied Science, University of Wisconsin-Milwaukee April, 2017

Multi-Model image registration Challenges and a novel approach to the problem Manifold Learning Fixed Image Moving Image Manifold Alignment Registration Algorithm Spatial Transformation Parameters Construct Hi-D Space Figure 2. Multi-modal image registration pipeline Figure 1. Multi-modal images of brain and their variations in intensities, (left) CT, (Middle) T1-weighted MR and (right) T2-weighted MR

Laplacian Eigenmap for structural representaion Challenges and a novel approach to the problem Figure 1. Multi-modal images of brain and their variations in intensities, (left) CT, (Middle) T1-weighted MR and (right) T2-weighted MR Figure 3. Structural representation with Laplacian Eigenmap

Results (a) T1-Weighted (b) T2-Weighted (c) PD-Weighted (d) CT Scan Figure 4. (Top) Original CT/MR scans of human brain; (Bottom) Structural representation of same scans. (a) T1-Weighted (b) T2-Weighted (c) PD-Weighted (d) CT Scan Representation Structural Brain Scans Original

Results (a) CT-T1 (b) CT-T2 (c) CT-PD (d) CT-T2 (20°rot.) Figure 5. Pairwise display of CT/MR registration. (Top) Unregistered scans, (Middle) Conventional registration results, (Bottom) Registration with the proposed method. (a) CT-T1 (b) CT-T2 (c) CT-PD (d) CT-T2 (20°rot.) Original Scans Registered Scans (Conventional) (proposed method) Results

Metric: Mutual Information Fixed Image Moving Image MI (Raw data) (Conv. method) (Proposed method) P001 - CT T1 0.8607 1.1352 1.1721 T1_rectified 0.8529 1.1919 1.1972 T2 0.8060 1.0269 1.0476 T2_rectified 0.8615 1.0665 1.1362 PD 0.8946 1.1222 1.1653 T1_30Rot 0.7976 1.1698 1.1774 T2_30Rot 0.7388 1.0379 1.0634 P002 – CT 0.8819 1.0660 1.1741 0.8473 1.1582 1.2013 0.7687 1.0257 1.1250 0.8056 1.0826 1.1854 Fixed Image Moving Image MI (Raw data) (Conv. method) (Proposed method) P002 – CT PD 0.8276 1.1034 1.2407 T1_30Rot 0.8205 1.0728 1.1891 T2_30Rot 0.7148 1.0089 1.1298 P003 - CT T1 0.6865 1.0424 1.0581 T2 0.7108 1.0551 1.0698 T2_rectified 0.7372 1.1009 1.1291 0.6535 1.0445 1.0516 P003 – T1 1.0732 1.0747 1.0973 P101 – CT 0.9422 1.2251 1.3003 0.9304 1.1612 1.2460 0.9940 1.3090 1.3817 MI is based on the statistical relationship between both volumes to be registered. Mutual information (MI) has become the established intensity similarity measure in multimodal registration because it accommodates different intensities between the modalities provided that they are relatively consistent within each modality The mutual information of images A and B measures the degree of dependence of A and B as the distance between the joint distribution and the distribution associated to the case of complete independence MI estimates the statistical dependence between corresponding voxel intensities, which is assumed to be maximal when the images are correctly aligned

Results Full data: T2-Weighted MR Full data: P001- CT Scan Partial data: T1-Weighted MR Full data: P001- CT Scan Partial data: P001- T1-Weighted MR Figure 6. Pairwise display of multi-modal partial data registration. Each pair: (Left) Unregistered scans, (Right) Registered scans by employing proposed method.