Use of Modified Masood Score with Prediction of Dominant Features to Classify Breast Lesions

  • N. M. SonaliEmail author
  • S. K. Padma
  • N. M. Nandini
  • T. S. Rekha
Part of the Studies in Computational Intelligence book series (SCI, volume 771)


In this chapter, we propose a novel method to improve the diagnostic accuracy for breast lesions based on certain cellular and nuclear features which become the criteria for final diagnosis. In the medical field, the accuracy of the diagnosis affects proper treatment of the condition. To overcome the problem of interobserver variability, a method of scoring is used to grade the lesions considered for the study. We have used the modified Masood score and designed an algorithm which classifies the various grades of breast lesions. We have used three classifiers: adaptive boosting, decision trees and random forests. These classifiers are effective tools for classifying breast lesions into benign, intermediate and malignant conditions. Principal component analysis using covariance and correlation is used to improve the diagnostic accuracy and for taking only the dominant features sufficient for classification.


Modified Masood score Adaptive boosting Decision tree Random forest Principal component analysis Covariance Correlation 


  1. 1.
  2. 2.
    Michael, J. 2002. Three decades of research in computer applications in health care. Journal of the American Medical Informatics Association 9 (2): 144–160.CrossRefGoogle Scholar
  3. 3.
    Wolberg, W.H., and O.L. Mangasrin. 1990. Multisurface method of pattern separation for medical diagnosis applied to breast cytology. Proceedings of the National Academy of Sciences of the United States of America 87 (23): 9193–9196.CrossRefGoogle Scholar
  4. 4.
    Takiar, R., D. Nadayil, and A. Nandakumar. 2010. Projections of number of cancer cases in India (2010–2020) by cancer groups. Asian Pacific Journal of Cancer Prevention 11 (4): 1045–1049.Google Scholar
  5. 5.
    Masood, Shahla. 2005. Cytomorphology of fibrocystic change, high-risk proliferative breast disease and premalignant breast lesions. Clinics in Laboratory Medicine 25 (4): 713–731.CrossRefGoogle Scholar
  6. 6.
    Mridha, Asit Ranjan, V.K. Iyer, K. Kapila, and K. Verma. 2006. Value of scoring system in classification of proliferative breast disease on fine needle aspiration cytology. Indian Journal of Pathology & Microbiology Journal 49 (3): 334–340.Google Scholar
  7. 7.
    Nandini, N.M., T.S. Rekha, and G.V. Manjunath. 2011. Evaluation of scoring system in cytological diagnosis and management of breast lesion with review of literature. Indian Journal of Cancer 48 (2): 240–245.CrossRefGoogle Scholar
  8. 8.
    Sheeba, D., and Chitrakala Sugumar. 2016. Palpable breast lesons-cytomorphological analysis and scoring system with histopathological correlation. IOSR-JDMS 15 (10): 25–29.CrossRefGoogle Scholar
  9. 9.
    Cherath, Smrithi Krishna, and Savithri Moothiringode Chithrabhaum. 2017. Evaluation of Masood’s and modified Masood’s scoring system in the cytological diagnosis of palpable breast lump aspirates. Journal of Clinical and Diagnostic Research 11 (4): EC06–EC10.Google Scholar
  10. 10.
    Freund, Yoav, and E. Robert Schapire. 1999. A short introduction to boosting. Journal of Japanese Society for Artificial Intelligence 14 (5): 771–780.Google Scholar
  11. 11.
    Quinlan, J.R. 1986. Induction of decision trees, vol. 1, 81–86. Kluwer Academic Publishers.Google Scholar
  12. 12.
    Breiman, L. 2001. Random forests 45 (1): 5–32.Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  • N. M. Sonali
    • 1
    Email author
  • S. K. Padma
    • 1
  • N. M. Nandini
    • 2
  • T. S. Rekha
    • 2
  1. 1.Department of Information Science and EngineeringSri Jayachamarajendra College of EngineeringMysoreIndia
  2. 2.Department of PathologyJSS Medical College, JSS UniversityMysoreIndia

Personalised recommendations