Evaluation of a Content-Based Image Retrieval Computer-Aided Diagnosis System for Breast Ultrasound Images Through Distance Similarity Measures

  • Min-jeong Kim
  • Hyun-chong ChoEmail author
Original Article



A content-based image retrieval (CBIR) computer-aided diagnosis (CADx) system using breast masses in ultrasound images has been developed and evaluated to assist radiologists with characterization processes. The purpose of this study is to improve the accuracy of breast cancer diagnoses by analyzing images and providing quantitative information to radiologists through the CADx system.


Two morphological features and six texture features of breast masses were extracted to design how the CADx system retrieves a mass similar to a query mass in a reference library. Based on extracted features from breast masses, the CADx system retrieves masses which are similar to the query mass from the reference library using a k-nearest neighbor (k-NN) method. To evaluate the CBIR CADx system, 39 similarity measures (nine similarity families, F0F8) based on the distance similarity were used. A receiver operating characteristic (ROC) analysis was conducted to evaluate the performance of the distance similarity measures.


The F0 family (Mahalanobis distance) measure used with the k-NN classifier provided slightly higher performance for the classification of malignant and benign masses as compared to those with the F1F8 family measures.


Breast cancer CADx system Medical image processing Ultrasound images 



This work was supported by the National Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) (no. 2017R1E1A1A03070297). This research was supported by the Ministry of Science and ICT (MSIT), Korea, under the Information Technology Research Center (ITRC) support program (IITP-2018-0-01433) supervised by the Institute for Information and Communications Technology Promotion (IITP).


  1. 1.
    Jung K-W, Won Y-J, Oh C-M, Kong H-J, Lee DH, Lee KH (2017) Prediction of cancer incidence and mortality in Korea, 2017. Cancer Res Treat 49:306–312CrossRefGoogle Scholar
  2. 2.
    Siegel RL, Miller KD, Jemal A (2017) Cancer statistics, 2017. CA Cancer J Clin 67:7–30CrossRefGoogle Scholar
  3. 3.
    Cho H, Hadjiiski L, Sahiner B, Chan HP, Paramagul C, Helvie M et al (2012) Interactive content-based image retrieval (CBIR) computer-aided diagnosis (CADx) system for ultrasound breast masses using relevance feedback. In: SPIE, medical imaging 2012, p 831509Google Scholar
  4. 4.
    Espín-López P, Martellosio A, Pasian M, Bozzi M, Perregrini L, Mazzanti A et al (2017) Breast cancer imaging at mm-waves: feasibility study on the safety exposure limits. In: Microwave conference (EuMC), 2016 46th European, pp 667–670Google Scholar
  5. 5.
    Hong AS, Rosen EL, Soo MS, Baker JA (2005) BI-RADS for sonography: positive and negative predictive values of sonographic features. Am J Roentgenol 184:1260–1265CrossRefGoogle Scholar
  6. 6.
    Cho H, Hadjiiski L, Sahiner B, Chan HP, Helvie M, Paramagul C et al (2011) Similarity evaluation in a content-based image retrieval (CBIR) CADx system for characterization of breast masses on ultrasound images. Med Phys 38:1820–1831CrossRefGoogle Scholar
  7. 7.
    Cui J, Sahiner B, Chan HP, Nees A, Paramagul C, Hadjiiski LM et al (2009) A new automated method for the segmentation and characterization of breast masses on ultrasound images. Med Phys 36:1553–1565CrossRefGoogle Scholar
  8. 8.
    Haralick RM, Shanmugam K, Dinstein I (1973) Texture features for image classification. IEEE Trans Syst Man Cybern SMC-3:610–621CrossRefGoogle Scholar
  9. 9.
    Belattar K, Mostefai S (2015) Similarity measures for content-based dermoscopic image retrieval: a comparative study. In: 2015 First international conference on new technologies of information and communication (NTIC), pp 1–6Google Scholar
  10. 10.
    Bo D, Zhangguan L, Cuixiao L (2015) An algorithm of image matching based on mahalanobis distance and weighted KNN graph. In: 2015 2nd international conference on information science and control engineering, Shanghai, 2015, pp 116–121Google Scholar
  11. 11.
    Bouhmala N (2016) How good is the euclidean distance metric for the clustering problem. In: 2016 5th IIAI international congress on advanced applied informatics (IIAI-AAI), pp 312–315Google Scholar
  12. 12.
    Cha SH (2007) Comprehensive survey on distance/similarity measures between probability density functions. Int J Math Model Meth Appl Sci 1(4):300–307Google Scholar

Copyright information

© The Korean Institute of Electrical Engineers 2019

Authors and Affiliations

  1. 1.Department of Electronic Engineering and Interdisciplinary Graduate Program for BIT Medical ConvergenceKangwon National UniversityChuncheonSouth Korea
  2. 2.Interdisciplinary Graduate Program for BIT Medical ConvergenceKangwon National UniversityChuncheonSouth Korea

Personalised recommendations