A review of breast tissue classification in mammograms

Part of: Research in Applied Computation Symposium (RACS 2011)


  • Gensheng Zhang
  • Wei Wang
  • Jucheol Moon
  • Jeong K Pack
  • Soon Ik Jeon


For women in the US breast cancer is the most commonly diagnosed cancer besides skin cancer and has become one of the major health issues in recent decades. Early detection through screening is one of key factors to reduce the death rates. The strong correlation between abnormality of breast tissues presented in mammograms and breast cancer shows that radiologists could benefit from Computer-Aided Diagnosis (CAD) systems with abilities of automated breast tissueclassification. This paper reviews recent advances in classification technologies of breast tissues. The major contribution of this paper is that we extensively discuss recent breast tissue classification technologies and compare three different types of approaches. According to our survey, we found that machine learning approaches could be chosen as an appropriate classification technology for a CAD system, considering efficiency and compatibility.