Computer Science - Research and Development

, Volume 26, Issue 1–2, pp 107–116 | Cite as

Model-based segmentation of pediatric and adult joints for orthopedic measurements in digital radiographs of the lower limbs

  • André GooßenEmail author
  • Eugen Hermann
  • Georg Martin Weber
  • Thorsten Gernoth
  • Thomas Pralow
  • Rolf-Rainer Grigat
Special Issue Paper


The growth of human bones forms a major problem when automatically segmenting orthopedic radiographs. Any template-based segmentation methods fails to fully capture these non-linear developments. However to extract orthopedic measurements or the bone age for patients of arbitrary age it is mandatory to have a segmentation scheme that deals with growth related changes. In this paper we propose a robust method based on Active Shape Models (ASMs) that on the one hand is invariant against the patient’s age and on the other hand generalizes well over the large inter-patient variability. Our method achieves an accuracy of 0.48 mm for adult patients and 0.64 mm for children on a large test set of 180 images, with the patient’s age covering a high range from less than one month to 93 years.


Segmentation Active shape models X-ray Digital radiography Orthopedics Bone age assessment 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Adelson E, Anderson CH, Bergen JR, Burt PJ, Ogden JM (1984) Pyramid methods in image processing. RCA Eng 29(6):33–41 Google Scholar
  2. 2.
    Behiels G, Maes F, Vandermeulen D, Suetens P (2002) Evaluation of image features and search strategies for segmentation of bone structures in radiographs using active shape models. Med Image Anal 6(1):47–62 CrossRefGoogle Scholar
  3. 3.
    Boewer M, Arndt H, Ostermann PW, Petersein J, Mutze S (2005) Length and angle measurements of the lower extremity in digital composite overview images. Eur Radiol 15(1):158–164 CrossRefGoogle Scholar
  4. 4.
    Bosch J, Mitchell S, Lelieveldt B, Nijland F, Kamp O, Sonka M, Reiber J (2002) Automatic segmentation of echocardiographic sequences by active appearance motion models. IEEE Trans Med Imaging 21(11):1374–1383 CrossRefGoogle Scholar
  5. 5.
    Boukala N, Favier E, Laget B, Radeva P (2004) Active shape model based segmentation of bone structures in hip radiographs. In: IEEE international conference on industrial technology, pp 1682–1687 Google Scholar
  6. 6.
    Chen Y, Ee X, Leow WK, Howe TS (2005) Automatic extraction of femur contours from hip X-ray images. In: Computer vision for biomedical image applications. Springer, Berlin, pp 200–209 CrossRefGoogle Scholar
  7. 7.
    Cootes TF, Taylor CJ (1992) Active shape models—‘smart snakes’. In: British machine vision conference. Springer, Berlin, pp 266–275 Google Scholar
  8. 8.
    Cootes TF, Taylor CJ (2001) Statistical models of appearance for medical image analysis and computer vision. In: Medical imaging, San Jose. SPIE, Bellingham, pp 236–248 Google Scholar
  9. 9.
    Cootes TF, Taylor CJ, Cooper DH, Graham J (1995) Active shape models—their training and application. Comput Vis Image Underst 61(1):38–59 CrossRefGoogle Scholar
  10. 10.
    Cootes TF, Edwards G, Taylor CJ (1999) Comparing active shape models with active appearance models. In: British machine vision conference. BMVA Press, Birmingham, pp 173–182 Google Scholar
  11. 11.
    Cootes TF, Baldock ER, Graham J (2000) An introduction to active shape models. Oxford University Press, London, pp 223–248 Google Scholar
  12. 12.
    Cootes TF, Wheeler GV, Walker KN, Taylor CJ (2002) View-based active appearance models. Image Vis Comput 20(9–10):657–664 CrossRefGoogle Scholar
  13. 13.
    Ding F, Leow WK, Howe TS (2007) Automatic segmentation of femur bones in anterior-posterior pelvis X-ray images. In: Computer analysis of images and patterns. Springer, Berlin, pp 205–212 CrossRefGoogle Scholar
  14. 14.
    Dong X, Zheng G (2008) Automatic extraction of femur contours from calibrated X-ray images: a Bayesian inference approach. In: 5th IEEE international symposium on biomedical imaging. Springer, Berlin, pp 57–60 Google Scholar
  15. 15.
    van Ginneken B, Frangi AF, Staal JJ, ter Haar Romeny BM, Viergever MA (2002) Active shape model segmentation with optimal features. IEEE Trans Med Imaging 21(8):924–933 CrossRefGoogle Scholar
  16. 16.
    Gooßen A, Schlüter M, Pralow T, Grigat RR (2008) A stitching algorithm for automatic registration of digital radiographs. In: Campilho A, Kamel M (eds) Image analysis and recognition, vol 5112. Springer, Berlin, pp 854–862 CrossRefGoogle Scholar
  17. 17.
    Gooßen A, Peters D, Gernoth T, Pralow T, Grigat RR (2009) Intelligent feature selection for model-based bone segmentation in digital radiographs. In: 9th international conference on technology and applications in biomedicine, ITAB 2009. IEEE Press, New York, pp 1–4 Google Scholar
  18. 18.
    Gower J (1975) Generalized procrustes analysis. Psychometrika 40:33–51 zbMATHCrossRefMathSciNetGoogle Scholar
  19. 19.
    Gregory J, Testi D, Stewart A, Undrill P, Reid D, Aspden R (2004) A method for assessment of the shape of the proximal femur and its relationship to osteoporotic hip fracture. Osteoporos Int 15:5–11 CrossRefGoogle Scholar
  20. 20.
    Hankemeier S, Gosling T, Richter M, Hufner T, Hochhausen C, Krettek C (2006) Computer-assisted analysis of lower limb geometry: higher intraobserver reliability compared to conventional method. Comput Aided Surg 11:81–86 CrossRefGoogle Scholar
  21. 21.
    Jeong Y, Radke RJ, Lovelock DM (2010) Bilinear models for inter- and intra-patient variation of the prostate. Phys Med Biol 55(13):3725 CrossRefGoogle Scholar
  22. 22.
    King DG, Steventon DM, O’Sullivan MP, Cook AM, Hornsby VPL, Jefferson IG, King PR (1994) Reproducibility of bone ages when performed by radiology registrars: an audit of Tanner and Whitehouse II versus Greulich and Pyle methods. Br J Radiol 67:848–851 CrossRefGoogle Scholar
  23. 23.
    Oost E, Koning G, Sonka M, Reiber JH, Lelieveldt BP (2005) Automated segmentation of X-ray left ventricular angiograms using multi-view active appearance models and dynamic programming. In: Frangi A, Radeva P, Santos A, Hernandez M (eds) Functional imaging and modeling of the heart. Springer, Berlin, pp 23–32 CrossRefGoogle Scholar
  24. 24.
    Pafilas D, Nayagam S (2008) The pelvic support osteotomy: indications and preoperative planning. Strateg Trauma Limb Reconstr 3(2):83–92 CrossRefGoogle Scholar
  25. 25.
    Paley D, Herzenberg J (2002) Principles of deformity correction, 1st edn. Springer, Berlin Google Scholar
  26. 26.
    Pyle SI, Hoerr NL (1955) Radiographic Atlas of the skeletal development of the knee, a standard of reference. Blackwell Scientific, Oxford Google Scholar
  27. 27.
    Rogers M, Graham J (2002) Robust active shape model search. In: Computer vision, ECCV 2002. Springer, Berlin, pp 289–312 Google Scholar
  28. 28.
    Romdhani S, Gong S, Psarrou A, Psarrou R (1999) A multi-view nonlinear Active Shape Model using Kernel PCA. In: British machine vision conference. BMVA Press, Birmingham, pp 483–492 Google Scholar
  29. 29.
    Ruppertshofen H, Lorenz C, Beyerlein P, Salah Z, Rose G, Schramm H (2010) Fully automatic model creation for object localization utilizing the generalized hough transform. In: Meinzer HP, Deserno TM, Handels H, Tolxdorff T (eds) Bildverarbeitung für die Medizin 2010: Algorithmen – Systeme – Anwendungen, pp 331–335 Google Scholar
  30. 30.
    Sabharwal S, Zhao C, McKeon J, McClemens E, Edgar M, Behrens F (2006) Computed radiographic measurement of limb-length discrepancy. J Bone Joint Surg 88:2243–2251 CrossRefGoogle Scholar
  31. 31.
    Schmitt H, Kappel H, Moser M, Cardenas-Montemayor E, Engelleiter K, Kuni B, Clarius M (2008) Determining knee joint alignment using digital photographs. Knee Surg Sports Traumatol Arthrosc 16:776–780 CrossRefGoogle Scholar
  32. 32.
    Schreiber J, Schubert R, Kuhn V (2006) Femur detection in radiographs using template-based registration. In: Handels H, Ehrhardt J, Horsch A, Meinzer HP, Tolxdorff T (eds) Bildverarbeitung für die Medizin 2006: Algorithmen – Systeme – Anwendungen, pp 111–115 CrossRefGoogle Scholar
  33. 33.
    Seise M, McKenna SJ, Ricketts IW, Wigderowitz CA (2005) Segmenting tibia and femur from knee X-ray images. In: Medical image understanding and analysis, pp 103–106 Google Scholar
  34. 34.
    Tenenbaum JB, Freeman WT (2000) Separating style and content with bilinear models. Neural Comput 12(6):1247–1283 CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2010

Authors and Affiliations

  • André Gooßen
    • 1
    Email author
  • Eugen Hermann
    • 2
  • Georg Martin Weber
    • 2
  • Thorsten Gernoth
    • 1
  • Thomas Pralow
    • 2
  • Rolf-Rainer Grigat
    • 1
  1. 1.Vision SystemsHamburg University of TechnologyHamburgDeutschland
  2. 2.Diagnostic X-RayPhilips HealthcareHamburgDeutschland

Personalised recommendations