Skip to main content
SpringerLink
Log in
Book cover

Guide to Medical Image Analysis pp 333–378Cite as

Detection and Segmentation by Shape and Appearance

  • Chapter

  • 4394 Accesses

Part of the book series: Advances in Computer Vision and Pattern Recognition ((ACVPR))

Abstract

Object detection in medical image analysis can be modeled as a search for an object model in the image. The model describes attributes such as shape and the appearance of the object. The search consists of fitting instances of the model to the data. A quality-of-fit measure determines whether one or several objects have been found.

Generating the model for a structure of interest can be difficult. It has to include knowledge about the acceptable variation of attributes within an object class while remaining suitably discriminative.

Several techniques to generate and use object models will be presented in this chapter. Information about acceptable object variation in these models is either generated from training or it is part of the model prototype.

This is a preview of subscription content, log in via an institution.

Chapter
USD   29.95
Price excludes VAT (USA)
  • Available as PDF
  • Read on any device
  • Instant download
  • Own it forever
eBook
USD   59.99
Price excludes VAT (USA)
  • Available as EPUB and PDF
  • Read on any device
  • Instant download
  • Own it forever
Softcover Book
USD   79.99
Price excludes VAT (USA)
  • Compact, lightweight edition
  • Dispatched in 3 to 5 business days
  • Free shipping worldwide - see info

Tax calculation will be finalised at checkout

Purchases are for personal use only

Learn about institutional subscriptions

Notes

  1. 1.

    The notation is the usual shorthand notation for HOM operators combining the two structuring elements in a single operator. The ‘0’s represent the erosion structuring element that is applied to the inverted binary image and the ‘1’ represent the erosion structuring element that is applied to the original binary image.

  2. 2.

    A rule of thumb borrowed from statistical pattern recognition for estimating likelihood functions from samples predict for a 100-dimensional feature space that at least 2100 samples would be needed.

  3. 3.

    Each of the models presented here can be essentially extended such that it can replace any of the other models. However, each of the models has been built with some idea about the objects to be described and the way necessary knowledge is to be gathered. It works most efficiently if objects or scenes to which it is applied follow this idea.

  4. 4.

    A FEM may be defined in a similar fashion than the mass-spring model by letting 1D springs being the elements. In such case, a bounded 2D or 3D object may be represented by a dense mesh of springs restricting shape variation.

  5. 5.

    Node values and force values of all nodes of an element (and later of the complete FEM mesh) are combined in a single vector. Hence, values for a node with index i in 2D have indices 2i and 2i+1 in the displacement vector u and the external force vector f.

  6. 6.

    This kind of assemblage becomes costly for the sparse matrix K if N is large. Faster methods to carry out this operation exist but the operation itself stays the same. For application in medical image analysis, however, the size of N is usually small (i.e., N≪100.000) and model creation does not happen often.

  7. 7.

    A dynamic system can be modeled without damping but damping prevents oscillation.

  8. 8.

    These modes are not contained in an ASM, since influence from rotation and translation is removed during normalization of the training data.

References

  • Al-Zubi S, Toennies KD (2003) Generalizing the active shape model by integrating structural knowledge to recognize hand drawn sketches. In: Proc CAIP 2003. LNCS, vol 2756, pp 320–328

    Google Scholar 

  • Bardinet E, Cohen LD, Ayache N (2003) Tracking medical 3D data with a parametric deformable model. In: Proc IEEE intl symp computer vision, pp 299–304

    Google Scholar 

  • Barr AH (1992) Rigid physically based superquadrics. In: Kirk D (ed) Graphics gems III. Academic Press, San Diego, pp 137–159

    Google Scholar 

  • Bergner S, Al-Zubi S, Toennies KD (2004) Deformable structural models. In: Proc IEEE intl conf image processing ICIP, pp 1875–1878

    Google Scholar 

  • Biederman I (1985) Human image understanding: recent research and a theory. Comput Vis Graph Image Process 32:29–73

    Article  Google Scholar 

  • Binford T (1987) Generalized cylinder representation. In: Encyclopedia of artificial intelligence. Wiley, New York, pp 321–323

    Google Scholar 

  • Blum H (1967) A transformation for extracting new descriptors of shape. In: Proc symp models for the perception of speech and visual form, pp 362–380

    Google Scholar 

  • Boykov Y, Veksler O, Zabih R (2001) Fast approximate energy minimization via graph cuts. IEEE Trans Pattern Anal Mach Intell 23(11):1222–1239

    Article  Google Scholar 

  • Brett AD, Taylor CJ (1999) A framework for automated landmark generation for automated 3d statistical model construction. In: Proc 16th intl conf information processing in medical imaging IPMI’99. LNCS, vol 1613, pp 376–381

    Google Scholar 

  • Chan T, Zhu W (2005) Level set based shape prior segmentation. In: Intl conf computer vision and pattern recognition, pp 1164–1170

    Google Scholar 

  • Cheung KW, Yeung DY, Chin RT (2002) On deformable models for visual pattern recognition. Pattern Recognit 35(7):1507–1526

    Article  MATH  Google Scholar 

  • Chevalier L, Jaillet F, Baskurt A (2001) 3D shape coding with superquadrics. In: Proc IEEE intl conf image processing ICIP, II, pp 93–96

    Google Scholar 

  • Cootes TF, Taylor CJ (1992) Active shape models—‘smart snakes’. In: Proc British machine vision conference.

    Google Scholar 

  • Cootes TF, Taylor CJ (1995) Combining point distribution models with shape models based on finite-element analysis. Image Vis Comput 13(5):403–409

    Article  Google Scholar 

  • Cootes TF, Taylor CJ, Cooper DH, Graham J (1995) Active shape models—their training and application. Comput Vis Image Underst 61(1):38–59

    Article  Google Scholar 

  • Cootes TF, Edwards GJ, Taylor CJ (1998) Active appearance models. In: 5th European conference on computer vision, ECCV1998. LNCS, vol 1407, pp 484–498

    Google Scholar 

  • Cremers D, Osher SJ, Soatto S (2006) Kernel density estimation and intrinsic alignment for shape priors in level set segmentation. Int J Comput Vis 69(3):335–351

    Article  Google Scholar 

  • Davies ER (1998) A modified Hough scheme for general circle location. Pattern Recognit 7(1):37–43

    Google Scholar 

  • Delingette H, Hebert M, Ikeuchi K (1992) Shape representation and image segmentation using deformable surfaces. Image Vis Comput 10(3):132–145

    Article  Google Scholar 

  • Dornheim L, Toennies KD, Dornheim J (2005) Stable dyaminc 3d shape models. In: IEEE intl conf image processing ICIP, III, pp 1276–1279

    Google Scholar 

  • Edelman S (1997) Computational theories in object recognition. Trends Cogn Sci 1:296–304

    Article  Google Scholar 

  • Engel K, Toennies KD (2008) Segmentation of the midbrain in transcranial sonographies using a two-component deformable model. In: 12th ann conf medical image understanding and analysis, pp 3–7

    Google Scholar 

  • Engel K, Toennies KD (2009) Hierarchical vibrations: a structural decomposition approach for image analysis. In: Energy minimization methods in computer vision and pattern recognition. LNCS, vol 5681, pp 317–330

    Chapter  Google Scholar 

  • Engel K, Toennies KD (2010) Hierarchical vibrations for part-based recognition of complex objects. Pattern Recognit 43(8):2681–2691

    Article  MATH  Google Scholar 

  • Engel K, Toennies KD, Brechmann A (2011) Part-based localisation and segmentation of landmark-related auditory cortical regions. Pattern Recognit 44(9):2017–2033

    Article  Google Scholar 

  • Ferrant M, Macq B, Nabavi A, Warfield SK (2000) Deformable modeling for characterizing biomedical shape changes. In: 9th intl conf discrete geometry for computer imagery, DGCI 2000. LNCS, vol 1953, pp 235–248

    Chapter  Google Scholar 

  • Frangi AF, Rueckert D, Schnabel J, Niessen WJ (2001) Automatic 3d ASM construction via atlas-based landmarking and volumetric elastic registration. In: Proc 17th intl conf information processing in medical imaging, IPMI 2001. LNCS, vol 2082, pp 78–91

    Google Scholar 

  • Freedman D, Zhang T (2005) Interactive graph cut based segmentation with shape priors. In: IEEE comp society conf on computer vision and pattern recognition (CVPR 2005), vol 1, pp 755–762

    Chapter  Google Scholar 

  • Giblin P, Kimia BB (2004) A formal classification of 3d medial axis points and their local geometry. IEEE Trans Pattern Recogn Mach Intell 26(2):238–251

    Article  Google Scholar 

  • Gloger O, Toennies KD, Kuehn JP (2011) Fully automatic liver volumetry using 3d level set segmentation for differentiated liver tissue types in multiple contrast MR datasets. In: Image analysis. LNCS, vol 6688, pp 512–523

    Chapter  Google Scholar 

  • Gong L, Pathak SD, Haynor DR, Cho PS, Kim Y (2004) Parametric shape modeling using deformable superellipses for prostate segmentation. IEEE Trans Med Imaging 23(3):340–349

    Article  Google Scholar 

  • Hamarneh G, McInerney T, Terzopoulos D (2001) Deformable organisms for automatic medical image analysis. In: Medical image computing and computer-assisted intervention, MICCAI 2001. LNCS, vol 2208, pp 66–76

    Chapter  Google Scholar 

  • Heimann T, Wolf I, Meinzer HP (2006) Active shape models for a fully automated 3d segmentation of the liver—an evaluation on clinical data. In: Medical image computing and computer-assisted intervention, MICCAI 2006. LNCS, vol 4191, pp 41–48

    Chapter  Google Scholar 

  • Jackway PT, Deriche M (1996) Scale-space properties of the multiscale morphological dilation-erosion. IEEE Trans Pattern Anal Mach Intell 18(1):38–51

    Article  Google Scholar 

  • Joshi S, Pizer SM, Fletcher PT, Yushkevich P, Thall A, Marron JS (2002) Multiscale deformable model segmentation and statistical shape analysis using medial descriptions. IEEE Trans Med Imaging 21(5):538–550

    Article  Google Scholar 

  • Kassim AA, Tan T, Tan KH (1999) A comparative study of efficient generalised Hough transform techniques. Image Vis Comput 17(10):737–748

    Article  Google Scholar 

  • Kichenassamy S, Kumar A, Olver P, Tannenbaum A, Yezzi A (1995) Gradient flows and geometric active contour models. In: 5th intl conf computer vision (ICCV’95), pp 810–817

    Google Scholar 

  • Lam L, Lee SW, Suen CY (1992) Thinning methodologies—a comprehensive survey. IEEE Trans Pattern Anal Mach Intell 14(9):869–885

    Article  Google Scholar 

  • Lindeberg T (1994) Scale-space theory: a basic tool for analysing structures at different scales. J Appl Stat 21(2):225–270

    Google Scholar 

  • Mandal C, Vemuri BC, Qin H (1998) A new dynamic FEM-based subdivision surface model for shape recovery and tracking in medical images. In: Medical image computing and computer-assisted intervention, MICCAI’98. LNCS, vol 1496, pp 753–760

    Google Scholar 

  • Marr D (1983) Vision. Henry Holt & Company, New York

    Google Scholar 

  • McInerney T, Terzopoulos D (1996) Deformable models in medical image analysis. Med Image Anal 1(2):91–108

    Article  Google Scholar 

  • Mokhtarian F, Mackworth A (1986) Scale-based description and recognition of planar curves and two-dimensional objects. IEEE Trans Pattern Anal Mach Intell 8(1):34–43

    Article  Google Scholar 

  • Okada T, Shimada R, Sato Y, Hori M, Yokota K, Nakamoto M, Chen YW, Nakamura H, Tamura S (2007) Automated segmentation of the liver from 3d CT images using probabilistic atlas and multi-level statistical shape model. In: Medical image computing and computer-assisted intervention, MICCAI 2007. LNCS, vol 4791, pp 86–93

    Chapter  Google Scholar 

  • Paloc C, Bello F, Kitney R, Darzi A (2002) Online multiresolution volumetric mass spring model for real time soft tissue deformation. In: Proc 5th intl conf medical image computing and computer-assisted intervention, MICCAI 2002. LNCS, vol 2489, pp 219–226

    Chapter  Google Scholar 

  • Pentland AP, Sclaroff S (1991) Closed-form solutions for physically-based modeling and reconstruction. IEEE Trans Pattern Anal Mach Intell 13(7):715–729

    Article  Google Scholar 

  • Petyt M (1998) Introduction to finite element vibration analysis. Cambridge University Press, Cambridge

    Google Scholar 

  • Pizer SM, Oliver WR, Bloomberg SH (1987) Hierarchical shape description via the multiresolution symmetric axis transforms. IEEE Trans Pattern Anal Mach Intell 9(4):505–511

    Article  Google Scholar 

  • Pizer SM, Fritsch DS, Yushkevich PA, Johnson VE, Chaney EL (1999) Segmentation, registration, and measurement of shape variation via image object shape. IEEE Trans Med Imaging 18(10):851–865

    Article  Google Scholar 

  • Provot X (1995) Deformation constraints in a mass model to describe rigid cloth behaviour. In: Graphics interface, pp 147–154

    Google Scholar 

  • Riesenhuber M, Poggio T (2000) Models of object recognition. Nat Neurosci Suppl 3:1190–1204

    Article  Google Scholar 

  • Rivlin E, Dickinson SJ, Rosenfeld A (1995) Recognition by functional parts. Comput Vis Image Underst 62(2):164–176

    Article  MATH  Google Scholar 

  • Sclaroff S, Pentland AP (1995) Modal matching for correspondence and recognition. IEEE Trans Pattern Anal Mach Intell 17(6):545–561

    Article  Google Scholar 

  • Terzopoulos D, Platt J, Barr A, Fleischer K (1987) Elastically deformable models. Proc SIGGRAPH Comput Graph 21(4):205–214

    Article  Google Scholar 

  • Terzopoulos D, Fleischer K (1988) Deformable models. Vis Comput 4(6):306–331

    Article  Google Scholar 

  • Terzopoulos D, Metaxas D (1991) Dynamic 3D models with local and global deformations: deformable superquadrics. IEEE Trans Pattern Anal Mach Intell 13(7):703–714

    Article  Google Scholar 

  • Toussaint GT (1978) The use of context in pattern recognition. Pattern Recognit 10(3):189–204

    Article  MathSciNet  MATH  Google Scholar 

  • Vu N, Manjunath BS (2008) Shape prior segmentation of multiple objects with graph cuts. In: IEEE comp society conf on computer vision and pattern recognition (CVPR 2008), pp 1–8

    Google Scholar 

  • Xu C, Prince JL (1998) Snakes, shapes, and gradient vector flow. IEEE Trans Image Process 7(3):359–369

    Article  MathSciNet  MATH  Google Scholar 

  • Zienkiewics OC, Taylor RL, Zhu JZ (2005) The finite element method: its basis & fundamentals, 6th edn. Elsevier, Amsterdam

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Computer Science Department, ISG, Otto-von-Guericke-Universität Magdeburg, Magdeburg, Germany

    Prof. Klaus D. Toennies

Authors
  1. Prof. Klaus D. Toennies
    View author publications

    You can also search for this author in PubMed Google Scholar

Rights and permissions

Reprints and permissions

Copyright information

© 2012 Springer-Verlag London Limited

About this chapter

Cite this chapter

Toennies, K.D. (2012). Detection and Segmentation by Shape and Appearance. In: Guide to Medical Image Analysis. Advances in Computer Vision and Pattern Recognition. Springer, London. https://doi.org/10.1007/978-1-4471-2751-2_11

Download citation

  • DOI: https://doi.org/10.1007/978-1-4471-2751-2_11

  • Publisher Name: Springer, London

  • Print ISBN: 978-1-4471-2750-5

  • Online ISBN: 978-1-4471-2751-2

  • eBook Packages: Computer ScienceComputer Science (R0)

Publish with us

Policies and ethics

Search

Navigation