Registration Algorithms for MR-Mammography

This page contains references and short descriptions of algorithms for the intra-visit registration of contrast-enhanced MR breast images. Please email me about any algorithm I've missed. No comments are given for references marked with * since I couldn't get hold of them.

M. H. Davis, A. Khotanzad, (Southern Methodist Univeristy, Dallas, USA)
D. P. Flamig (Los Alamos National Laboratories, Los Alamos, USA),
S. E. Harms (University of Arkansas for Medical Sciences, Little Rock, USA)

• Davis95 [1]*
• Davis97 [2]
  AIM: to find realistic transformation for given correspondance of anatomic landmarks
  METHODS: elastic body spline, physical model of homogenous, isotropic 3D elastic body, actual control point pairing by block matching method based on regional normalized correlation,
  TESTS: tested on MRI of breast without contrast agent, simulated deformations (2nd order Taylor series warp) and random selected landmarks, results assessed based on image similarity (mean squared error, mean absolute error, correlation coefficient), comparison of elastic body spline to thin plate spline and volume spline, test of influence of number of control-points and type of initial transformation (none, rigid, affine)

M. S. NessAiver, T. Krebs, J. Wong, B. S. Fernandes, M. Steverson (University of Maryland at Baltimore)

• NessAiver95 [3]*
• NessAiver96 [4]
  METHODS: individual rigid transformation per 2D slice, minimize standard deviation of subtracted images
  TESTS: standard deviation of subtracted images, visual quality of subtracted images

R. Kumar, J. C. Asmuth, K. Hanna, J. Bergen (David Sarnoff Research Center, Princton, USA)
C. Hulka, D. B. Kopans, R. Weisskoff, R. Moore (Massachusetts General Hospital, Boston, USA)

• Kumar95 [5]*
• Kumar96 [6]
  METHODS: global 3D affine followed by one local flow iteration, minize sum of squares of differences, coarse to fine strategy with 3D pyramid
  TESTS: visual inspection of subtracted images and in cine mode

C. S. Zuo, B. L. Buff, T. G. Mahon, T. Z. Wong (Deaconess Hosptial, Boston, USA),
A. Jiang (Massachusetts General Hosptial and Harvard Medical School, Boston, USA)

• Zuo96 [7]
  METHODS: rigid transformation for rectangular subvolume enclosing the breast, minimize variance of ratio images of the two image volumes,
  TESTS: variance across breast boundary for subtracted images, visually better subtracted images, SNR within lesion improves, registration of simulated image pair in correspondance (simulated lesion without motion) produced rotations angles of less than 0.05$^o$ and translations less than 0.02mm

B. A. Porter, J. P. Smith, (First Hill Diagnostic Imaging, Seattle, USA),
P. Margosian (Picker International Cleveland, USA),
J. V. Hanjal, (Hammersmith Hospital, London, UK)

• Porter96 [8]
  METHODS: translation by hand to one voxel accuracy
  TESTS: visual inspection

P. Undrill, L. Vieira, T. W. Redpath, F. J. Gilbert (University of Aberdeen, Aberdeen, UK)

• Undrill96 [9]
  METHODS: a) constrained principle axis determination: determine translation + rotation + scaling from principle components of general moments of 3D object; b) correlation in image domain: translation + rotation + scaling, minimize global variance between corresponding voxels, c) polynomial warping determined through corresponding (identified) control points
  TESTS: visual inspection of two cases
• Vieira99 [10]
  METHODS: a) automatic image registration (AIR): minimise variance of ratio, warping polynomials of 2nd to 5th order, b) spatial parametric mapping (SPM): sum of squares of differences, linear combination of basis function (3D discrete cosine transformations)
  TESTS: simulated deformations like a shear parallel to rib cage, no contrast, qualitative analysis: only intensity changes assessed not target registration error

S. Huwer, J. Rahmel, A. v. Wagenheim (University of Kaiserslautern, Kaiserslautern, Germany)

• Huwer96 [11]
  METHODS: modified Kohonen network model, local non-affine transformation, intensity difference, gradient, relevant set of pixel
  TESTS: visual inspection of one case

M. Davey, I. D. Wilkinson, F. Balen, H. Mumtaz, M. Paley, D. L. Plummer, M. A. Hall-Craggs, W. R. Lees, A. Linney (University College London; The Middlesex Hospital, London, UK)

• Davey97 [12]
  METHODS: rigid transformation per breast, root mean square error mininimization
  TESTS: visually, quality of subtracted images

P. M. Hayton, M. Brady, L. Tarassenko, S. M. Smith, N. Moore ( Oxford University, The John Radcliffe Hospital, Oxford, UK)

• Hayton97 [13],
  METHODS: minimize residual error when fitting pharma-cokinetic curve, optic flow, 2D
  TESTS: residual error, registration error from inducing global translation, visual inspection of image
• Hayton99 [14]
  METHODS: local mutual information, Bayes probabilities, optic flow field, B-spline fit, smoothing regularizator, multi-scale, correct for bias field
  TESTS: results from 3 patients, visual inspection, consistency of forward and backward transformation after induced 20deg rotation of a none aligned image pair

S. Rueckert, C. Tanner, J. A. Schnabel, L. I. Sonoda, A. D. Castellano-Smith, D. L. Hill, D. J. Hawkes (King's College, London, UK)
E. R. E. Denton, S. C. Rankin (Guy's Hospital, London, UK)
A. Degenhard, C. Hayes, M. O. Leach, (Institute of Cancer Research and the Royal Marsden NHS Trust, Sutton, UK)
D. R. Hose (University of Sheffield, Sheffield, UK)

• Rueckert98-99 [15]-[17],
  METHODS: rigid or affine registration followed by a free-form deformation (FFD), FFD is based on a regularly spaced grid of control points, multi-resolution approach, normalized mutual information as similarity measure
  TESTS: comparision to other similarity measures, quality of subtracted images
• Denton99 [18]
  AIM: evaluation of registration performance
  METHODS: ranking of quality of subtracted images before and after rigid, affine and non-rigid registration by blinded radiologists
• Tanner00 [19]
  METHODS: local volume preservation of registration by coupling of control points
  TESTS: analysis of volume change for 15 enhancing lesions
• Schnabel01,03 [20,21]
  METHODS: validation method of non-rigid registration by applying physically plausible breast deformations generated by a finite element model to image pairs where no motion was visible
  TESTS: demonstration of method
• Tanner01 [22]
  AIM: comparison of biomechanical models base on finite element methods to justify plausibility of simulated breast deformations
  METHODS: MR images before and after deformation of a volunteers breast, optimal surface displacement values from 3D non-rigid registration, obtain solution of finite element model
  TESTS: error of predicting displacement of internal breast structures on 12 landmarks
• Schnabel02 [23],Tanner02 [24]
  AIM: validation of non-rigid registration for MR-mammography
  METHODS: based on simulated breast deformations using finite element models (Schnabel03 [21]) for a multi-level registration approach (Schnabel02 [25]) and a volume-preserving multi-resolution registration approach

C. Meyer, J. Boes, B. Kim, P. Bland (University of Michigan, Ann Arbor, USA)

• Meyer98 [26]
  AIM: want to find minimal set of of control points
  METHODS: initial estimate from manually selected corresponding control points, move these around randomly (up to 3mm) to maximize global mutual information
  TESTS: compare curvature of cost function and variance of final positions of control points, visual inspection of two patients

H. Fischer, M. Otte, C. Ehritt-Braun, M. Büchert, P. Peschl, J. Hennig (University of Freiburg)

• Fischer98 [27]
  METHODS: subdivided into equally sized subvolumes, connect local transformations by trilinear Berzier splines, optimized using linear correlation or linear entropy coefficient
  TESTS: visual inspection
• Fischer00 [28]

T. Brückner, M. V. Knopp (German Cancer Research Center, Heidelberg, Germany)
R. Lucht, G. Brix (Federal Office for Radiation Protection, Neuherberg, Germany)

• Lucht98 [29]
  METHODS: automatic feature extraction(edges from median filter and opening), find corresponding features (translations, feature overlap, bidding system), bi-linear interpolation
  TESTS: feature overlap
• Lucht00 [30]
  METHODS: pre-alignment for any data with a shift greater than 1 cm, feature extraction (edges from median filter and opening), find corresponding features, smooth transformation by Gaussian filtering, 2D only
  TESTS: assessed on correlation coefficient, visual inspection of animation, tracing anatomical landmarks in continous film sequences (20 patients)
• Brückner00 [31]
  AIM: evaluation of performance of rigid and elastic matching algorithms; evaluation of mutual information as matching criterion
  METHODS: a) 2D rigid registration, maximise mutual information; b) elastic matching as described in Lucht00 [30]
  TESTS: assessed within breast region using mutual information, visual inspection in cine mode by a blinded reader for 10 patients

S. Krishnan, T. L. Chenevert, M. A. Helvie, F. L. Londy (University of Michigan Hospitals, Ann Arbor, USA)

• Krishnan99 [32]
  AIM: to correct the phase shifts due to patient motion in k-space data prior to offline keyhole reconstruction
  METHODS: in-vivo, registrations accounting for translations
  TESTS: on phantom with induced translations and Gd-DTPA, 64 patients, quantitatively on subtraction edge artifacts

M. A. Wirth (RMIT University, Melbourne, Australia)

• Wirth99 [33]
  METHODS: 2D registration, a) affine transformation, b) radial basis functions (RBF) incorporating thin plate splines, c) multiquadric RBF; control points automatic from breast contour and manually from inner region
  TESTS: three 2D slices, 96 control points, correlation coefficient, sum of squares differences, difference images

J. R. Reichenbach, J. Hopfe, W. A. Kaiser (Friedrich-Schiller-University, Jena, Germany)
M. E. Bellemann (University of Applied Sciences, Jena, Germany)
R. Lucht (Federal Office for Radiation Protection, Neuherberg, Germany)

• Reichenbach00 [34]
  METHODS:image split into single breast image, mutual information as image similarity measure, global translation, individual 2D rotation for each slice,
  TESTS: three volunteers, mutual information plots, visual inspection
• Reichenbach02 [35]
  METHODS: image automatically split into single breast image, mutual information as similarity measure, rotation parameter for per-selected equidistant slices
  TESTS: on translated and rotated phantom, volunteers repositioned, no contrast agent, performance assessed using mutual information which was also used as optimization criterion

T. Rohlfing, C. R. Maurer (Stanford University Medical Centre, STanford, USA)

• Rohlfing01 [36]
  METHODS: volume preserving non-rigid registration, global rigid registration and free-form deformation with regularly spaced control-points, cost-function consists of two optimazation criteria: normalized mutual information and average absolute logarithm of the local volume change, multi-level B-splines for non-uniform local refinement of control-point spacing
  TESTS: volume change, visual inspection of one case

B. Fischer, J. Modersitzki (University of Lübeck, Lübeck, Germany)

• Fischer02 [37,38]
  METHODS: curvature type regularization term, sum of squares differences
  TESTS: illustrated on 2D MR-mammograms, reduction of sum of squares differences, visual inspection

Bibliography

1

M. H. Davis, A. Khotanzad, D. P. Flamig, and S. E. Harms, ``Coordinate Transformation in 3D Image Matching by a Physical Based Method-Elastic Body Spline,'' in Int. Symposium on Computer Vision, Coral Gables, Florida, p. 218, 1995.

2

M. H. Davis, A. Khotanzad, D. P. Flemig, and S. E. Harms, ``A Physics-Based Coordinate Transformation for 3-D Image Matching,'' IEEE Transactions on Medical Imaging, vol. 16, no. 3, pp. 317-328, 1997.

3

M. S. NessAiver, B. S. Fernandes, T. Krebs, J. Wong, and M. Steverson, ``Three Dimensional Subtraction of Pre- and Post-contrast Breast Image with Translational and Rotational Registration,'' in Proceedings of the Third Scientific Meeting of the International Society for MR in Medicine, Nice, France, p. 438, 1995.

4

M. S. NessAiver, T. Krebs, and J. Wong, ``Improved Registration for Subtration of Pre- and Post-Contrast 3D Breast Images,'' in Proceedings of the Fourth Scientific Meeting of the International Society for MR in Medicine, New York, USA, p. 38, 1996.

5

R. Kumar, K. Hanna, J. C. Asmuth, J. Bergen, C. Hulka, D. B. Kopans, R. Weisskoff, and R. Moore, ``Detecting Lesions in Magnetic Resonance Breast Scans,'' in AIPR Workshop on Tools and Techniques for Modeling and Simulation, Washington DC, USA, Vol. 2645, pp. 181-190, 1995.

6

R. Kumar, J. C. Asmuth, K. Hanna, J. Bergen, C. Hulka, D. B. Kopans, R. Weisskoff, and R. Moore, ``Application of 3D Registration for Detecting Lesions in Magnetic Resonance Breast Scans,'' in Proceedings SPIE Medical Imaging 1996, Image Processing, San Diego, CA, pp. 646-656, 1996.

7

C. S. Zuo, A. Jiang, B. L. Buff, T. G. Mahon, and T. Z. Wong, ``Automatic Motion Correction for Breast MR Imaging,'' Radiology, vol. 198, no. 3, pp. 903-906, 1996.

8

B. A. Porter, J. P. Smith, P. Margosian, and J. V. Hanjal, ``Image Re-registration for MR Subtraction Techniques,'' in Proceedings of the Fourth Scientific Meeting of the International Society for MR in Medicine, New York, USA, p. 760, 1996.

9

P. E. Undrill, T. W. Redpath, and F. J. Gilbert, ``Reduction of Movement Artefacts in Comparative 3D Magnetic Resonance (MR) Breast Imagaging,'' in Proceedings SPIE Medical Imaging 1996, Image Processing, San Diego, CA, pp. 922-930, 1996.

10

L. Vieira and P. Undrill, ``Use of a Breast Model to Test the Performance of Image Registration Methods,'' in Medical Image Understanding and Analysis, Oxford, UK, pp. 181-184, 1999.

11

S. Huwer, J. Rahmel, and A. v. Wagenheim, ``Data-Driven Registration for Local Deformations,'' Pattern Recognition Letters, vol. 17, pp. 951-957, 1996.

12

M. Davey, I. D. Wilkinson, F. Balen, H. Mumtaz, M. Paley, D. L. Plummer, M. A. Hall-Craggs, W. R. Lees, and A. Linney, ``Detecting Breast Cancer: Registration of Pre- and Post-contrast T1-weighted 3D Data Sets,'' in Proceedings of the Fifth Scientific Meeting of the International Society for MR in Medicine, Vancouver, Canada, p. 1048, 1997.

13

P. M. Hayton, M. Brady, L. Tarassenko, and N. Moore, ``Analysis of Dynamic MR Breast Images using a Model of Contrast Enhancement,'' Medical Image Analysis, vol. 1, pp. 207-224, September 1997.

14

P. M. Hayton, M. Brady, S. M. Smith, and N. Moore, ``A non-rigid Registration Algorithm for Dynamic Breast MRI Images,'' Artificial Intelligence, vol. 114, pp. 125-156, 1999.

15

D. Rueckert, C. Hayes, C. Studholme, P. Summers, M. Leach, and D. J. Hawkes, ``Non-rigid Registration of Breast MR Images using Mutual Information,'' in Medical Image Computing and Computer-Assisted Intervention, Cambridge, USA, pp. 1144-1152, 1998.

16

D. Rueckert, L. I. Sonoda, E. Denton, S. Rankin, C. Hayes, M. O. Leach, D. L. Hill, and D. J. Hawkes, ``Comparison and Evalution of Rigid and Non-rigid Registration of Breast MR Images,'' in Proceedings SPIE Medical Imaging 1999, Image Processing, San Diego, CA, 1999.

17

D. Rueckert, L. I. Sonoda, C. Hayes, D. L. Hill, M. O. Leach, and D. J. Hawkes, ``Non-rigid Registration using Free-Form Deformations: Application to Breast MR Images,'' IEEE Transactions on Medical Imaging, vol. 18(8), pp. 712-721, 1999.

18

E. R. E. Denton, L. I. Sonoda, D. Rueckert, S. C. Rankin, C. Hayes, M. O. Leach, and D. J. Hawkes, ``Comparison and Evaluation of Rigid, Affine, and Nonrigid Registration of Breast MR Images,'' Journal of Computer Assisted Tomography, vol. 23(5), pp. 800-805, 1999.

19

C. Tanner, J. A. Schnabel, D. Chung, M. J. Clarkson, D. Rueckert, D. L. G. Hill, and D. J. Hawkes, ``Volume and Shape Preservation of Enhancing Lesions when Applying Non-rigid Registration to a Time Series of Contrast Enhanced MR Breast Images,'' in Medical Image Computing and Computer-Assisted Intervention, Pittsburgh, USA, Vol. 1935 of Lecture Notes in Computer Science, Springer Verlag, pp. 327-337, 2000.

20

J. A. Schnabel, D. Rueckert, M. Quist, J. M. Blackall, A. Castellano-Smith, T. Hartkens, G. P. Penney, W. A. Hall, C. L. Truwit, F. A. Gerritsen, D. L. G. Hill, and D. J. Hawkes, ``A Generic Framework for Non-Rigid Registration based on Non-Uniform Multi-Level Free-Form Deformations,'' in Medical Image Computing and Computer-Assisted Intervention, Utrecht, Netherlands, Vol. 2208 of Lecture Notes in Computer Science, Springer Verlag, pp. 573-581, 2001.

21

J. A. Schnabel, C. Tanner, A. D. Castellano-Smith, A. Degenhard, M. O. Leach, D. R. Hose, D. L. G. Hill, and D. J. Hawkes, ``Validation of Non-Rigid Image Registration using Finite Element Methods: Application to Breast MR Images,'' IEEE Transactions on Medical Imaging, vol. 22(1), 2003. In press.

22

C. Tanner, A. Degenhard, J. A. Schnabel, A. Castellano-Smith, C. Hayes, L. I. Sonoda, M. O. Leach, D. R. Hose, D. L. G. Hill, and D. J. Hawkes, ``A Method for the Comparison of Biomechanical Breast Models,'' in IEEE Workshop on Mathematical Methods in Biomedical Image Analysis, Kauai, USA, pp. 11-18, 2001.

23

J. A. Schnabel, C. Tanner, A. D. Castellano-Smith, A. Degenhard, C. Hayes, M. O. Leach, D. R. Hose, D. L. G. Hill, and D. J. Hawkes, ``Finite Element Based Validation of Non-Rigid Registration using Single- and Multi-Level Free-Form Deformations: Application to Contrast-Enhanced MR Mammography,'' in Proceedings SPIE Medical Imaging 2002, Image Processing, San Diego, CA, pp. 550-581, 2002.

24

C. Tanner, J. A. Schnabel, A. Degenhard, A. Castellano-Smith, C. Hayes, M. O. Leach, D. R. Hose, D. L. G. Hill, and D. J. Hawkes, ``Validation of Volume-Preserving Non-Rigid Registration: Application to Contrast-Enhanced MR-Mammography.,'' in Medical Image Computing and Computer-Assisted Intervention, Tokyo, Japan, Vol. 2489 of Lecture Notes in Computer Science, Springer Verlag, pp. 307-314, 2002.

25

J. A. Schnabel, C. Tanner, A. D. Castellano-Smith, A. Degenhard, C. Hayes, M. O. Leach, D. R. Hose, D. L. G. Hill, and D. J. Hawkes, ``Validation of Non-Rigid Registration of Contrast-Enhanced MR Mammography using Finite Element Methods.,'' in In Proc. Workshop on Bildverarbeitung fuer die Medizin, Informatik Aktuell, pp. 143-146, 2002.

26

C. Meyer, J. Boes, B. Kim, and P. Bland, ``Evaluation of Control Point Selection in Automatic, Mutual Information Driven, 3D Warping,'' in Medical Image Computing and Computer-Assisted Intervention, Cambridge, USA, pp. 944-951, 1998.

27

H. Fischer, M. Otte, C. Ehritt-Braun, M. Buechert, P. Peschl, and J. Hennig, ``Local Elastic Motion Correction in MR-Mammography,'' in Proceedings of the Sixth Scientific Meeting of the International Society for MR in Medicine, Sydney, Australia, p. 725, 1998.

28

H. Fischer, ``Automatic Elastic Motion Correction in MR-Mammography,'' in 2nd International Congress on MR-Mammography, Jena, Germany, published in European Radiology 10:9, pp. F48-F49, 2000.

29

R. Lucht, M. V. Knopp, and G. Brix, ``Elastic Matching of Dynamic MR Image Data Sets of the Female Breast,'' in Proceedings of the Sixth Scientific Meeting of the International Society for MR in Medicine, Sydney, Australia, p. 940, 1998.

30

R. Lucht, M. V. Knopp, and G. Brix, ``Elastic Matching of Dynamic MR Mammographic Images,'' Magnetic Resonance in Medicine, vol. 43, pp. 9-16, January 2000.

31

T. Brückner, R. Lucht, and G. Brix, ``Comparison of Rigid and Elastic Matching of Dynamic Magnetic Resonance Mammographic Images by Mutual Information,'' Medical Physics, vol. 27, pp. 2456-2461, October 2000.

32

S. Krishnan, T. L. Chenevert, M. A. Helvie, and F. L. Londy, ``Linear Motion Correction in Three Dimensions Applied to Dynamic Gadolinium Enhanced Breast Imaging,'' Medical Physics, vol. 26, pp. 707-714, 1999.

33

M. A. Wirth, A Nonrigid Approach to Medical Image Registration: Matching Images of the Breast.
PhD thesis, RMIT University Melbourne, 1999.

34

J. R. Reichenbach, J. Hopfe, R. Lucht, W. A. Kaiser, and M. E. Bellemann, ``Image Registration of Serial 3D MR Breast Data Using Local Parameters,'' in Proceedings of the Eighth Scientific Meeting of the International Society for MR in Medicine, Denver, USA, p. 2173, 2000.

35

J. R. Reichenbach, J. Hopfe, M. E. Bellemann, and W. A. Kaiser, ``Development and Validation of an Algorithm for Registration of Serial 3D MR Breast Data Sets,'' Magnetic Resonance Materials in Physics, Biologie and Medicine, vol. 14, pp. 249-258, 2002.

36

T. Rohlfing and C. R. Maurer, ``Intensity-Based Non-rigid Registration Using Adaptive Multilevel Free-Form Deformation with an Incompressibility Constraint,'' in Medical Image Computing and Computer-Assisted Intervention, Utrecht, Netherlands, Vol. 2208 of Lecture Notes in Computer Science, Springer Verlag, pp. 111-119, 2001.

37

B. Fischer and J. Modersitzki, ``Fast Curvature Based Registration of MR-Mammography Images,'' in In Proc. Workshop on Bildverarbeitung fuer die Medizin, Informatik Aktuell, pp. 139-143, 2002.

38

B. Fischer and J. Modersitzki, ``A Unified Approach to Fast Image Registration and a New Curvature Based Registration Technique,'' Technical Report, Preprint A-02-07, Institute of Mathematics, Medical University of Luebeck, pp. 1-17, 2002.