Computational Radiology Laboratory Harvard Medical School www.crl.med.harvard.edu Children’s Hospital Department of Radiology Boston Massachusetts A Survey.

Slides:



Advertisements
Similar presentations
Yinyin Yuan and Chang-Tsun Li Computer Science Department
Advertisements

Gordon Wright & Marie de Guzman 15 December 2010 Co-registration & Spatial Normalisation.
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,
© Fraunhofer MEVIS Toward Automated Validation of Sketch-based 3D Segmentation Editing Tools Frank Heckel 1, Momchil I. Ivanov 2, Jan H. Moltz 1, Horst.
PET/CT Working Group Update Jayashree Kalpathy-Cramer Sandy Napel.
Proportion Priors for Image Sequence Segmentation Claudia Nieuwenhuis, etc. ICCV 2013 Oral.
ProbExplorer: Uncertainty-guided Exploration and Editing of Probabilistic Medical Image Segmentation Ahmed Saad 1,2, Torsten Möller 1, and Ghassan Hamarneh.
Jeroen Hermans, Frederik Maes, Dirk Vandermeulen, Paul Suetens
Automatic Identification of Bacterial Types using Statistical Image Modeling Sigal Trattner, Dr. Hayit Greenspan, Prof. Shimon Abboud Department of Biomedical.
A UTOMATIC S EGMENTATION OF T HALAMIC N UCLEI WITH STEPS L ABEL F USION J.Su 1, T. Tourdias 2, M.Saranathan 1, and B.K.Rutt 1 1 Department of Radiology,
SEMANTIC FEATURE ANALYSIS IN RASTER MAPS Trevor Linton, University of Utah.
Multiple Criteria for Evaluating Land Cover Classification Algorithms Summary of a paper by R.S. DeFries and Jonathan Cheung-Wai Chan April, 2000 Remote.
HMM-BASED PATTERN DETECTION. Outline  Markov Process  Hidden Markov Models Elements Basic Problems Evaluation Optimization Training Implementation 2-D.
1 Learning to Detect Objects in Images via a Sparse, Part-Based Representation S. Agarwal, A. Awan and D. Roth IEEE Transactions on Pattern Analysis and.
Announcements  Project proposal is due on 03/11  Three seminars this Friday (EB 3105) Dealing with Indefinite Representations in Pattern Recognition.
Shape Modeling International 2007 – University of Utah, School of Computing Robust Smooth Feature Extraction from Point Clouds Joel Daniels ¹ Linh Ha ¹.
Interactive, GPU-Based Level Sets for 3D Segmentation Aaron Lefohn Joshua Cates Ross Whitaker University of Utah Aaron Lefohn Joshua Cates Ross Whitaker.
September 27, / 18 Automatic Segmentation of Neonatal Brain MRI Marcel Prastawa 1, John Gilmore 2, Weili Lin 3, Guido Gerig 1,2 University.
Yujun Guo Kent State University August PRESENTATION A Binarization Approach for CT-MR Registration Using Normalized Mutual Information.
12-Apr CSCE790T Medical Image Processing University of South Carolina Department of Computer Science 3D Active Shape Models Integrating Robust Edge.
Image Registration Narendhran Vijayakumar (Naren) 12/17/2007 Department of Electrical and Computer Engineering 1.
Recovering Articulated Object Models from 3D Range Data Dragomir Anguelov Daphne Koller Hoi-Cheung Pang Praveen Srinivasan Sebastian Thrun Computer Science.
Evaluating the Quality of Image Synthesis and Analysis Techniques Matthew O. Ward Computer Science Department Worcester Polytechnic Institute.
P. Rodríguez, R. Dosil, X. M. Pardo, V. Leborán Grupo de Visión Artificial Departamento de Electrónica e Computación Universidade de Santiago de Compostela.
Robust Linear Registration of CT images using Random Regression Forests Ender Konukoglu 1, Antonio Criminisi 1, Sayan Pathak 2, Duncan Robertson 1, Steve.
An Integrated Pose and Correspondence Approach to Image Matching Anand Rangarajan Image Processing and Analysis Group Departments of Electrical Engineering.
Li Wang1, Feng Shi1, Gang Li1, Weili Lin1, John H
Model-based Automatic AC/PC Detection on Three-dimensional MRI Scans Babak A. Ardekani, Ph.D., Alvin H. Bachman, Ph.D., Ali Tabesh, Ph.D. The Nathan S.
Computational Radiology Laboratory Harvard Medical School Brigham and Women’s Hospital Children’s Hospital Boston Massachusetts.
Computational Radiology Laboratory Harvard Medical School Children’s Hospital Department of Radiology Boston Massachusetts MRI.
Multi-Modal Quantitative Analysis of Pediatric Focal Epilepsy Andy Eow Medical Vision Group CSAIL, MIT.
DTU Medical Visionday May 27, 2009 Generative models for automated brain MRI segmentation Koen Van Leemput Athinoula A. Martinos Center for Biomedical.
Quality Assessment for LIDAR Point Cloud Registration using In-Situ Conjugate Features Jen-Yu Han 1, Hui-Ping Tserng 1, Chih-Ting Lin 2 1 Department of.
Enhanced Correspondence and Statistics for Structural Shape Analysis: Current Research Martin Styner Department of Computer Science and Psychiatry.
J OURNAL C LUB : Cardoso et al., University College London, UK “STEPS: Similarity and Truth Estimation for Propagated Segmentations and its application.
Medical Image Analysis Image Registration Figures come from the textbook: Medical Image Analysis, by Atam P. Dhawan, IEEE Press, 2003.
A Registration-Based Atlas Propagation Framework for Automatic Whole Heart Segmentation Xiahai Zhuang (PhD) Centre for Medical Image Computing University.
Copyright © 2010 Siemens Medical Solutions USA, Inc. All rights reserved. Hierarchical Segmentation and Identification of Thoracic Vertebra Using Learning-based.
Surgical Planning Laboratory Brigham and Women’s Hospital Boston, Massachusetts USA a teaching affiliate of Harvard Medical School Functional Data Analysis.
Chapter 4: CS6891 Computational Medical Imaging Analysis Chapter 4: Image Visualization Jun Zhang Laboratory for Computational Medical Imaging & Data Analysis.
Spatio-Temporal Free-Form Registration of Cardiac MR Image Sequences Antonios Perperidis s /02/2006.
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.
Prostate Cancer CAD Michael Feldman, MD, PhD Assistant Professor Pathology University Pennsylvania.
National Alliance for Medical Image Computing Segmentation Foundations Easy Segmentation –Tissue/Air (except bone in MR) –Bone in CT.
A B C D E F A ABSTRACT A novel, efficient, robust, feature-based algorithm is presented for intramodality and multimodality medical image registration.
KNN & Naïve Bayes Hongning Wang Today’s lecture Instance-based classifiers – k nearest neighbors – Non-parametric learning algorithm Model-based.
Spatial Smoothing and Multiple Comparisons Correction for Dummies Alexa Morcom, Matthew Brett Acknowledgements.
©2005 Surgical Planning Laboratory, ARR Slide 1 Prostate Image Processing Steven Haker, PhD.
Statistical Analysis An Introduction to MRI Physics and Analysis Michael Jay Schillaci, PhD Monday, April 7 th, 2007.
1 Chapter 8: Model Inference and Averaging Presented by Hui Fang.
MultiModality Registration Using Hilbert-Schmidt Estimators By: Srinivas Peddi Computer Integrated Surgery II April 6 th, 2001.
Bayesian Methods Will Penny and Guillaume Flandin Wellcome Department of Imaging Neuroscience, University College London, UK SPM Course, London, May 12.
CIVET seminar Presentation day: Presenter : Park, GilSoon.
Introduction to Medical Imaging Regis Introduction to Medical Imaging Registration Alexandre Kassel Course
A 2D/3D correspondence building method for reconstruction of a 3D bone surface model Longwei Fang
Automatic segmentation of brain structures
KNN & Naïve Bayes Hongning Wang
Ron Kikinis, M.D ‡. Wanmei Ou, MSc §, Polina Golland, Ph.D. §, William Wells III, Ph.D. ‡§, Carsten Richter ‡, Steven Pieper, Ph.D. ¥, Haiying Liu ‡, Wendy.
Validation and Evaluation of Algorithms
HST 583 fMRI DATA ANALYSIS AND ACQUISITION
Non-linear Realignment Using Minimum Deformation Averaging for Single-subject fMRI at Ultra-high Field Saskia Bollmann1, Steffen Bollmann1, Alexander.
Tensor-based Surface Modeling and Analysis
Moo K. Chung1,3, Kim M. Dalton3, Richard J. Davidson2,3
Multi-modality image registration using mutual information based on gradient vector flow Yujun Guo May 1,2006.
Computational Neuroanatomy for Dummies
Detecting Gray Matter Maturation via Tensor-based Surface Morphometry
Hierarchical Models and
Anatomical Measures John Ashburner
MultiModality Registration using Hilbert-Schmidt Estimators
MS-SEG Challenge: Results and Analysis on Testing Data
Presentation transcript:

Computational Radiology Laboratory Harvard Medical School Children’s Hospital Department of Radiology Boston Massachusetts A Survey of Validation Techniques for Image Segmentation and Registration, with a focus on the STAPLE algorithm Simon K. Warfield, Ph.D. Associate Professor of Radiology Harvard Medical School

Computational Radiology Laboratory. Slide 2 Outline Validation of image segmentation –Overview of approaches –STAPLE Validation of image registration STAPLE algorithm available as open source software from: – –

Computational Radiology Laboratory. Slide 3 Segmentation Goal: identify or label structures present in the image. Many methods: –Interactive or manual delineation, –Supervised approaches with user initialization, –Alignment with a template, –Statistical pattern recognition. Applications: –Quantitative measurement of volume, shape or location of structures, –Provides boundary for visualization by surface rendering. Newborn MRI Segmentation.

Computational Radiology Laboratory. Slide 4 Validation of Image Segmentation Spectrum of accuracy versus realism in reference standard. Digital phantoms. –Ground truth known accurately. –Not so realistic. Acquisitions and careful segmentation. –Some uncertainty in ground truth. –More realistic. Autopsy/histopathology. –Addresses pathology directly; resolution. Clinical data ? –Hard to know ground truth. –Most realistic model.

Computational Radiology Laboratory. Slide 5 Validation of Image Segmentation Comparison to digital and physical phantoms: –Excellent for testing the anatomy, noise and artifact which is modeled. –Typically lacks range of normal or pathological variability encountered in practice. MRI of brain phantom from Styner et al. IEEE TMI 2000

Computational Radiology Laboratory. Slide 6 Comparison To Higher Resolution MRIPhotographMRI Provided by Peter Ratiu and Florin Talos.

Computational Radiology Laboratory. Slide 7 Comparison To Higher Resolution PhotographMRI PhotographMicroscopy Provided by Peter Ratiu and Florin Talos.

Computational Radiology Laboratory. Slide 8 Comparison to Autopsy Data Neonate gyrification index –Ratio of length of cortical boundary to length of smooth contour enclosing brain surface

Computational Radiology Laboratory. Slide 9 Staging Stage 3 Stage 5 Stage 4 Stage 6 Stage 3: at 28 w GA shallow indentations of inf. frontal and sup. Temp. gyrus (1 infant at 30.6 w GA, normal range: 28.6 ± 0.5 w GA) Stage 4: at 30 w GA 2 indentations divide front. lobe into 3 areas, sup. temp.gyrus clearly detectable (3 infants, 30.6 w GA ± 0.4 w, normal range: 29.9 ± 0.3 w GA) Stage 5: at 32 w GA frontal lobe clearly divided into three parts: sup., middle and inf. Frontal gyrus (4 infants, 32.1 w GA ± 0.7 w, normal range: 31.6 ± 0.6 w GA) Stage 6: at 34 w GA temporal lobe clearly divided into 3 parts: sup., middle and inf. temporal gyrus (8 infants, 33.5 w GA ± 0.5 w normal range: 33.8 ± 0.7 w GA) “Assessment of cortical gyrus and sulcus formation using MR images in normal fetuses”, Abe S. et al., Prenatal Diagn 2003

Computational Radiology Laboratory. Slide 10 Neonate GI: MRI Vs Autopsy

Computational Radiology Laboratory. Slide 11 GI Increase Is Proportional to Change in Age.

Computational Radiology Laboratory. Slide 12 GI Versus Qualitative Staging

Computational Radiology Laboratory. Slide 13 Neonate Gyrification

Computational Radiology Laboratory. Slide 14 Validation of Image Segmentation Comparison to expert performance; to other algorithms. Why compare to experts ? –Experts are currently doing the segmentation tasks that we seek algorithms for. –Surgical planning. –Neuroscience research. What is the appropriate measure for such comparisons ?

Computational Radiology Laboratory. Slide 15 Measures of Expert Performance Repeated measures of volume –Intra-class correlation coefficient Spatial overlap –Jaccard: Area of intersection over union. –Dice: increased weight of intersection. –Vote counting: majority rule, etc. Boundary measures –Hausdorff, 95% Hausdorff. Bland-Altman methodology: –Requires a reference standard. Measures of correct classification rate: –Sensitivity, specificity ( Pr(D=1|T=1), Pr(D=0|T=0) ) –Positive predictive value and negative predictive value (posterior probabilities Pr(T=1|D=1), Pr(T=0|D=0) )

Computational Radiology Laboratory. Slide 16 Validation of Image Segmentation STAPLE (Simultaneous Truth and Performance Level Estimation): –An algorithm for estimating performance and ground truth from a collection of independent segmentations.

Computational Radiology Laboratory. Slide 17 STAPLE papers –Image segmentation with labels: Warfield, Zou, Wells ISBI 2002 Warfield, Zou, Wells MICCAI Warfield, Zou, Wells, IEEE TMI Commowick and Warfield IPMI 2009 –Image segmentation with boundaries: Warfield, Zou, Wells MICCAI Warfield, Zou, Wells PTRSA –Diffusion data and vector fields: Commowick and Warfield IEEE TMI 2009

Computational Radiology Laboratory. Slide 18 STAPLE: Estimation Problem Complete data density: Binary ground truth T i for each voxel i. Expert j makes segmentation decisions D ij. Expert performance characterized by sensitivity p and specificity q. –We observe expert decisions D. If we knew ground truth T, we could construct maximum likelihood estimates for each expert’s sensitivity (true positive fraction) and specificity (true negative fraction):

Computational Radiology Laboratory. Slide 19 Expectation-Maximization Since we don’t know ground truth T, treat T as a random variable, and solve for the expert performance parameters that maximize: Parameter values θ j =[p j q j ] T that maximize the conditional expectation of the log-likelihood function are found by iterating two steps: –E-step: Estimate probability of hidden ground truth T given a previous estimate of the expert quality parameters, and take expectation. –M-step: Estimate expert performance parameters by comparing D to the current estimate of T.

Computational Radiology Laboratory. Slide 20 Probability Estimate of True Labels Estimate probability of tissue class in reference standard:

Computational Radiology Laboratory. Slide 21 Binary Input: True Segmentation

Computational Radiology Laboratory. Slide 22 Expert Performance Estimate p (sensitivity, true positive fraction) : ratio of expert identified class 1 to total class 1 in the image. q (specificity, true negative fraction) : ratio of expert identified class 0 to total class 0 in the image.

Computational Radiology Laboratory. Slide 23 Newborn MRI Segmentation

Computational Radiology Laboratory. Slide 24 Newborn MRI Segmentation Summary of segmentation quality (posterior probability Pr(T=t|D=t) ) for each tissue type for repeated manual segmentations. Indicates limits of accuracy of interactive segmentation.

Computational Radiology Laboratory. Slide 25 Expert and Student Segmentations Test imageExpert consensusStudent 1 Student 2Student 3

Computational Radiology Laboratory. Slide 26 Phantom Segmentation ImageExpertStudentsVotingSTAPLE ImageExpert segmentation Student segmentations

Computational Radiology Laboratory. Slide 27 STAPLE Summary Key advantages of STAPLE: –Estimates ``true’’ segmentation. –Assesses expert performance. Principled mechanism which enables: –Comparison of different experts. –Comparison of algorithm and experts. Extensions for the future: –Prior distribution or extended models for expert performance characteristics. –Estimate bounds on parameters.

Computational Radiology Laboratory. Slide 28 Image registration A metric: measures similarity of images given an estimate of the transformation. Best metric depends on nature of the images. Alignment quality ultimately possible depends on model of transformation. The transformation is identified by solving an optimization problem. –Seek the transform parameters that maximize the metric of image similarity

Computational Radiology Laboratory. Slide 29 Validation of Registration Compare transformations –Take some images, apply a transformation to them. –Estimate the transform using registration –How well does the estimated transformation match the applied transform? Check alignment of key image features –Fiducial alignment –Spatial overlap Segment structures, assess overlap after alignment.

Computational Radiology Laboratory. Slide 30 Intraoperative Nonrigid Registration Fast: it should not take more than 1 min to make the registration. Robust: the registration should work with poor quality image, artifacts, tumor... Physics based: we are not only concerned in the intensity matching, but also interested in recovering the physical (mechanical) deformation of the brain. Accurate: neuro-surgery needs a precise knowledge of the position of the structures. Archip et al. NeuroImage 2007

Computational Radiology Laboratory. Slide 31 Block Matching Algorithm Divide a global optimization problem in many simple local ones Highly parallelizable, as blocks can be matched independently. Similarity measure: coefficient of correlation

Computational Radiology Laboratory. Slide 32 Block Matching Algorithm Displacement estimates are noisy.

Computational Radiology Laboratory. Slide 33 Patient-specific Biomechanical Model Pre-operative image Automatic brain segmentation Brain finite element model (linear elastic)

Computational Radiology Laboratory. Slide 34 Registration Validation Landmark matching assessment in six cases Parallel version runs in 35 seconds on a 10 dual 2GHz PC cluster –7x7x7 block size –11x11x25 window –1x1x1 step – blocks – tetrahedra 60 landmarks: –Average error = 0.75mm –Maximum error = 2.5mm –Data voxel size 0.8x0.8x2.5 mm 3

Computational Radiology Laboratory. Slide 35 Registration Validation 11 prospective consecutive cases, Alignment computed during the surgery. Estimate of the registration accuracy – 95% Hausdorff distance of the edges of the registered preoperative MRI and the intraoperative MRI.

Computational Radiology Laboratory. Slide 36 Automatic selection of fiducials (1)Non-rigid alignment of preoperative MPRAGE. Contours extracted from (1) with the Canny edge detector (2) Intraoperative whole brain SPGR at 0.5T Contours extracted from (2) with the Canny edge detector 95% Hausdorff metric computed

Computational Radiology Laboratory. Slide 37 Alignment improvement Tumor positionTumor pathology Non-rigid registration – preop to intraop scans (95% Hausdorff distance) Max Displacement measured (mm) Rigid registration accuracy – preop to intraop (mm) Non-Rigid registration accuracy – preop to intraop (mm) Ratio Rigid/Non- Rigid Case 1right posterior frontaloligoastrocytoma Grade II Case 2left posterior temporalglioblastoma Grade IV Case 3left medial temporalglioblastoma Grade IV Case 4left temporalanaplastic oligoastrocytoma Grade III Case 5right frontaloligoastrocytoma Grade II Case 6left frontalanaplastic astrocytoma Grade III Case 7right medial temporalanaplastic astrocytoma Grade III Case 8right frontaloligoastrocytoma Grade II Case 9right frontotemporaloligoastrocytoma Grade II Case 10right occipitalanaplastic oligodendroglioma Grade III Case 11left frontotemporal oligodendroglioma Grade II AVG

Computational Radiology Laboratory. Slide 38 Visualization of aligned data Matched preoperative fMRI and DT-MRI aligned with intraoperative MRI. Tensor alignment: Ruiz et al. 2000

Computational Radiology Laboratory. Slide 39 Conclusion Validation strategies for registration: –Comparison of transformations. –Fiducials Manual, automatic. –Overlap statistics – as for segmentation. Validation strategies for segmentation: –Digital and physical phantoms. –Comparison to domain experts. –STAPLE.

Computational Radiology Laboratory. Slide 40 Acknowledgements Neil Weisenfeld. Andrea Mewes. Richard Robertson. Joseph Madsen. Karol Miller. Michael Scott. This study was supported by: R01 RR021885, R01 EB008015, R01 GM Collaborators William Wells. Kelly H. Zou. Frank Duffy. Arne Hans. Olivier Commowick. Alexandra Golby. Vicente Grau.