Automatic ultrasound segmentation and morphology based diagnosis of solid breast tumors

  • Ruey-Feng Chang
  • Wen-Jie Wu
  • Woo Kyung Moon
  • Dar-Ren Chen


Ultrasound (US) is a useful diagnostic tool to distinguish benign from malignant masses of the breast. It is a very convenient and safe diagnostic method. However, there is a considerable overlap benignancy and malignancy in ultrasonic images and interpretation is subjective. A high performance breast tumors computer-aided diagnosis (CAD) system can provide an accurate and reliable diagnostic second opinion for physicians to distinguish benign breast lesions from malignant ones. The potential of sonographic texture analysis to improve breast tumor classifications has been demonstrated. However, the texture analysis is system-dependent. The disadvantages of these systems which use texture analysis to classify tumors are they usually perform well only in one specific ultrasound system. While Morphological based US diagnosis of breast tumor will take the advantage of nearly independent to either the setting of US system and different US machines. In this study, the tumors are segmented using the newly developed level set method at first and then six morphologic features are used to distinguish the benign and malignant cases. The support vector machine (SVM) is used to classify the tumors. There are 210 ultrasonic images of pathologically proven benign breast tumors from 120 patients and carcinomas from 90 patients in the ultrasonic image database. The database contains only one image from each patient. The ultrasonic images are captured at the largest diameter of the tumor. The images are collected consecutively from August 1, 1999 to May 31, 2000; the patients’ ages ranged from 18 to 64 years. Sonography is performed using an ATL HDI 3000 system with a L10-5 small part transducer. In the experiment, the accuracy of SVM with shape information for classifying malignancies is 90.95% (191/210), the sensitivity is 88.89% (80/90), the specificity is 92.5% (111/120), the positive predictive value is 89.89% (80/89), and the negative predictive value is 91.74% (111/121).


breast ultrasound computer-aided diagnosis level set shape support vector machine 


  1. Pandey N, Salcic Z, Sivaswamy J: Fuzzy logic based microcalcification detection. neural networks for signal processing X, 2000. In: Proceedings of the 2000 IEEE Signal Processing Society Workshop 2000, Sydney, NSW, Australia, pp 662–671.Google Scholar
  2. Bassett, LW, Liu, TH, Giuliano, AE, Gold, RH 1991The prevalence of carcinoma in palpable vs impalpable, mammographically detected lesionsAJR1572124PubMedGoogle Scholar
  3. Rubin, M, Horiuchi, K, Joy, N, Haun, W, Read, R, Ratzer, E, Fenoglio, M 1997Use of fine needle aspiration for solid breast lesions is accurate and cost-effectiveAm J Surg174694696PubMedCrossRefGoogle Scholar
  4. Gisvold, JJ, Martin, JK,Jr. 1984Prebiopsy localization of nonpalpable breast lesionsAJR Am J Roentgenol143477481PubMedGoogle Scholar
  5. Shankar, PM, Reid, JM, Ortega, H, Piccoli, CW, Goldberg, BB 1993Use of non-rayleigh statistics for identification of tumors in ultrasonic B-scans of the breastIEEE Trans Med Imag12687692CrossRefGoogle Scholar
  6. Stavros, AT, Thickman, D, Rapp, CL, Dennis, MA, Parker, SH, Sisney, GA 1995Solid breast nodules: use of sonography to distinguish between benign and malignant lesionsRadiology196123134PubMedGoogle Scholar
  7. Canny, J 1986A computational approach to edge detectionIEEE Trans Pattern Anal Machine Intell8679698CrossRefGoogle Scholar
  8. Pavlidis, T 1982Algorithms for Graphics and Image Processing.Computer Science PressRockville, MDGoogle Scholar
  9. Pavlidis, T, Horowitz, SL 1974Segmentation of plane curvesIEEE Trans Comput27860870CrossRefGoogle Scholar
  10. Osher, S, Sethian, J 1988Fronts propagating with curvature-dependent speed: Algorithms based on Hamilton–Jacobi equationsJ Comput Phys791249CrossRefGoogle Scholar
  11. Kass, M, Witkin, A, Terzopoulos, D 1988Snakes: active contour modelsInt J Comp Vis1321331CrossRefGoogle Scholar
  12. Corsi, C, Saracino, G, Sarti, A, Lamberti, C 2002Left ventricular volume estimation for real-time three-dimensional echocardiographyIEEE Trans Med Imaging2112021208PubMedCrossRefGoogle Scholar
  13. Wang X, Wee WG: A new deformable contour method. In: Image Analysis and Processing, 1999. Proceeding International Conference on 1999, Venice, Italy, pp 430–435.Google Scholar
  14. Chen YM, Thiruvenkadam S, Tagare HD, Hung F, Wilson D, Geiser EA: On the incorporation of shape priors into geometric active contours. In: Variational and Level Set Methods in Computer Vision, 2001. Proceedings. IEEE Workshop on 2001, Vancouver, BC, Canada, pp 145–152 Google Scholar
  15. Vapnik, VN 1995The Nature of Statistical Learning Theory.Springer-VerlagNew YorkGoogle Scholar
  16. Chapelle, O, Haffner, P, Vapnik, VN 1999Support vector machines for histogram-based image classificationIEEE Trans Neural Netw, IEEE Trans1010551064CrossRefGoogle Scholar
  17. Pontil, M, Verri, A 1998Support vector machines for 3D object recognitionIEEE Trans Pattern Anal and Machine Intell20637646CrossRefGoogle Scholar
  18. Gerig, G, Kubler, O, Kikinis, R, Jolesz, FA 1992Nonlinear anisotropic filtering of MRI dataIEEE Trans Med Imaging11221232PubMedCrossRefGoogle Scholar
  19. Perona, J, Malik, J 1990Scale-scape and edge-detection using anisotropic diffusionIEEE Trans Pattern Anal Machine Intell12629639CrossRefGoogle Scholar
  20. Czerwinski RN, Jones DL, O’Brien WD, Jr.: Edge detection in ultrasound speckle noise. Proc IEEE International Conference Image Processing, Austin, TX 1994, Austin, TX, pp 304–308.Google Scholar
  21. Czerwinski, RN, Jones, DL, O’Brien, WD 1999Detection of lines and boundaries in speckle images - Application to medical ultrasoundIEEE Trans Med Imaging18126136PubMedCrossRefGoogle Scholar
  22. Otsu, N 1979A threshold selection method form gray-level histogramsIEEE Trans Systems, Man, and Cybernetics96266CrossRefGoogle Scholar
  23. Sethian, JA 1996Level Set Methods: Evolving Interfaces in Geometry, Fluid Mechanics, Computer Vision, and Materials Science.Cambridge University PressCambridgeGoogle Scholar
  24. Chen, DR, Chang, RF, Huang, YL, Chou, YH, Tiu, CM, Tsai, PP 2000Texture analysis of breast tumors on sonogramsSemin Ultrasound CT MR21308316PubMedCrossRefGoogle Scholar
  25. Chen, DR, Chang, RF, Huang, YL 2000Breast cancer diagnosis using self-organizing map for sonographyUltrasound Med Biol26405411PubMedCrossRefGoogle Scholar
  26. Chen, DR, Chang, RF, Huang, YL 1999Computer-aided diagnosis applied to US of solid breast nodules by using neural networksRadiology213407412PubMedGoogle Scholar
  27. Chen, DR, Kuo, WJ, Chang, RF, Moon, WK, Lee, CC 2002Use of the bootstrap technique with small training sets for computer-aided diagnosis in breast ultrasoundUltrasound Med Biol28897902PubMedCrossRefGoogle Scholar
  28. McPherson, DD, Aylward, PE, Knosp, BM, Bean, JA, Kerber, RE, Collins, SM, Skorton, DJ 1986Ultrasound characterization of acute myocardial ischemia by quantitative texture analysisUltrason Imaging8227240PubMedCrossRefGoogle Scholar
  29. Garra, BS, Krasner, BH, Horii, SC, Ascher, S, Mun, SK, Zeman, RK 1993Improving the distinction between benign and malignant breast lesions: the value of sonographic texture analysisUltrason Imaging15267285PubMedCrossRefGoogle Scholar
  30. Goldberg, V, Manduca, A, Ewert, DL, Gisvold, JJ, Greenleaf, JF 1992Improvement in specificity of ultrasonography for diagnosis of breast tumors by means of artificial intelligenceMed Phys1914751481PubMedCrossRefGoogle Scholar
  31. Chen, CM, Chou, YH, Han, KC, Hung, GS, Tiu, CM, Chiou, HJ, Chiou, SY 2003Breast lesions on sonograms: computer-aided diagnosis with nearly setting-independent features and artificial neural networksRadiology226504514PubMedCrossRefGoogle Scholar
  32. Bertsekas, DB, Tsitsiklis, JN 1989Parallel and Distributed Computation: Numerical Methods.Prentice HallEnglewood Cliffs, NJGoogle Scholar

Copyright information

© Springer 2005

Authors and Affiliations

  • Ruey-Feng Chang
    • 1
  • Wen-Jie Wu
    • 1
  • Woo Kyung Moon
    • 2
  • Dar-Ren Chen
    • 3
  1. 1.Department of Computer Science and Information EngineeringNational Chung Cheng UniversityCjanghuaTaiwan
  2. 2.Department of Diagnostic RadiologySeoul National University HospitalSouth Korea
  3. 3.Department of SurgeryChanghua Christian HospitalChanghuaTaiwan

Personalised recommendations