The Evaluation of Effects on Breast Cancer Diagnoses When Using Mammographic Semantic Information

  • Authors:

  • Da Qi;Erika R. Denton;Joanna M. Leason;Diaa Othman;Reyer Zwiggelaar

  • Affiliations:

  • Department of Computer Science, Aberystwyth University, Aberystwyth, UK;Department of Radiology, Norfolk and Norwich University Hospital NHS Trust, Norwich, UK;Department of Radiology, Norfolk and Norwich University Hospital NHS Trust, Norwich, UK;Bronglais Hospital, Breast Surgery, Ceredigion and Mid-Wales NHS Trust, Aberystwyth, UK;Department of Computer Science, Aberystwyth University, Aberystwyth, UK

  • Venue:
  • IWDM '08 Proceedings of the 9th international workshop on Digital Mammography
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

In this paper, we describe the evaluation of the effects of mammographic semantic information in breast cancer diagnoses. A brief description of relations between semantic information and image features are given. We demonstrate the experiments based on mammographic semantic information and the MIAS database. Mammograms were annotated by expert radiologists with semantic information and assigned NHSBSP five-point score. Two classifiers were applied to automatically classify the mammogram into NHSBSP five-point score using the semantic information and radiologists also classified the mammograms by their own annotated semantic information. The analysis of the experimental results provides further understanding when using mammographic semantic information in breast cancer diagnosis. It also indicated a common knowledge base and links between computers and human experts.