A Robust Medical Image Recognition System Employing Edge-Based Feature Vector Representation

  • Masakazu Yagi
  • Tadashi Shibata
  • Chihiro Tanikawa
  • Kenji Takada
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2749)


The medical image recognition system based on a robust edge-based feature vector representation, called Projected Principal-Edge Distribution (PPED), has been developed. The robust nature of the PPED vector representation has been demonstrated in the cephalometric landmark identification, expert dentists’ practice. In contradistinction with the previous works for the optimization of the PPED, the feasibility of the system has been demonstrated by applying to 26 cephalometric landmarks with 250 head films and the ones with the degradation. The system has been successfully applied to not only the cephalometric landmarks on the hard tissue but also the ones on the soft tissue without tuning the algorithm. The robust nature of the system against the degradation in the X-ray films has also been demonstrated.


Recognition Performance Hard Tissue Template Generation Robust Nature European Signal Processing 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


  1. 1.
    Levy-Mandel, D., Venetsanopoulos, A. V., Tsotsos, J. K.: Knowledge-Based Landmarking of Cephalograms. COMPUTERS AND BIOMEDICAL RESEARCH 19, 1986, pp. 282–309.CrossRefGoogle Scholar
  2. 2.
    Rudolph, D. J., Sinclair, P. M., Coggins, J. M.: Automatic Computerized Radiographic Identification of Cephalometric Landmarks. American Journal of Orthodontics and Dentofacial Orthopedics, Feb. 1998;113, pp. 173–179.CrossRefGoogle Scholar
  3. 3.
    Yen-ting, C., Kuo-sheng, C., Ja-kuang, L.: Improving Cephalogram Analysis through Feature Subimage Extraction. IEEE Engineering in Medicine and Biology Magazine, Vol. 18, No. 1, 1999, pp. 25–31.CrossRefGoogle Scholar
  4. 4.
    Ogawa, M., Ito, K., Shibata, T.: A General-Purpose Vector-Quantization Processor employing Two-Dimensional Bit-Propagating Winner-Take-All. Digest of Technical Papers of 2002 Symposium on VLSI Circuits, p.p. 244–247, Honolulu, June 13–15, 2002.Google Scholar
  5. 5.
    Yamasaki, T., Shibata, T.: Analog Soft-Pattern-Matching Classifier Using Floating-Gate MOS Technology. 14th Conf. Neural Information Processing Systems: Natural and Synthetic, 2001 (NIPS*2001), pp. 1131–1138, Vancouver, Dec. 2001.Google Scholar
  6. 6.
    Yagi, M., Adachi, M., Shibata, T.: A Hardware-Friendly Soft-Computing Algorithm for Image Recognition. 10th European Signal Processing Conference (EUSIPCO 2000), pp. 729–732, Tampere, Finland, Sept. 4–8, 2000Google Scholar
  7. 7.
    Yagi, M., Shibata, T.: A Human-Perception-like Image Recognition System based on PAP Vector Representation with Multi Resolution Concept. 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2002), Vol. I, pp.1041–1048, Florida, May 13–17, 2002.Google Scholar
  8. 8.
    Yagi, M., Shibata, T., Takada, K.: Optimizing Feature-Vector Extraction Algorithm from Grayscale Images for Robust Medical Radiograph Analysis. Fourth International Conference on Multimedia and Image Processing (IFMIP 2002), Vol. 13, pp.251–257, Orland, June 9–13, 2002.Google Scholar
  9. 9.
    Yagi, M., Shibata, T., Takada, K.: Human-Perception-like Image Recognition System Based on the Associative Processor Architecture. XI European Signal Processing Conference (EUSIPCO2002), Vol. I, pp. 103–106, Sep. 3–6, 2002 Toulouse, France (2002).Google Scholar
  10. 10.
    Mostafa, Y. A., EL-Mangoury, N. H., Salah, A., Rasmy, E. M.: Automated Cephalographic Soft-Tissue Analysis. JCO / Sept. 1990, Vol. XXIV No. 9.Google Scholar
  11. 11.
    Yagi, M., Shibata, T.: An Image Representation Algorithm Compatible with Neural-Associative-Processor-Based Hardware Recognition Systems. IEEE Transactions on Neural Networks. (To be published)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Masakazu Yagi
    • 1
  • Tadashi Shibata
    • 2
  • Chihiro Tanikawa
    • 1
  • Kenji Takada
    • 1
  1. 1.Department of Orthodontics and Dentofacial Orthopedics, Graduate School of DentistryOsaka UniversitySuita, OsakaJapan
  2. 2.Department of Frontier Informatics, School of Frontier ScienceThe University of TokyoTokyoJapan

Personalised recommendations