Comparative study of ROC regression techniques-Applications for the computer-aided diagnostic system in breast cancer detection

  • Authors:

  • María Xosé Rodríguez-Álvarez;Pablo G. Tahoces;Carmen Cadarso-Suárez;María José Lado

  • Affiliations:

  • Unit of Biostatistics, Department of Statistics and Operations Research, University of Santiago de Compostela, Spain and Complexo Hospitalario Universitario de Santiago de Compostela (CHUS), Santi ...;Department of Electronics and Computer Science, University of Santiago de Compostela, Spain;Unit of Biostatistics, Department of Statistics and Operations Research, University of Santiago de Compostela, Spain and Instituto de Investigación Sanitaria de Santiago de Compostela (IDIS), ...;Department of Computer Science, University of Vigo, Spain

  • Venue:
  • Computational Statistics & Data Analysis
  • Year:
  • 2011

Quantified Score

Hi-index 0.03

Visualization

Abstract

The receiver operating characteristic (ROC) curve is the most widely used measure for statistically evaluating the discriminatory capacity of continuous biomarkers. It is well known that, in certain circumstances, the markers' discriminatory capacity can be affected by factors, and several ROC regression methodologies have been proposed to incorporate covariates in the ROC framework. An in-depth simulation study of different ROC regression models and their application in the emerging field of automatic detection of tumour masses is presented. In the simulation study different scenarios were considered and the models were compared to each other on the basis of the mean squared error criterion. The application of the reviewed ROC regression techniques in evaluating computer-aided diagnostic (CAD) schemes can become a major factor in the development of such systems in Radiology.