Title: Modelling semantics from image data: opportunities from LIDC

Authors: Daniela S. Raicu, Ekarin Varutbangkul, Jacob D. Furst, Samuel G. Armato III

Addresses: Intelligent Multimedia Processing Laboratory, School of Computer Science, Telecommunications, and Information Systems, DePaul University, Chicago, IL 60604, USA. ' Intelligent Multimedia Processing Laboratory, School of Computer Science, Telecommunications, and Information Systems, DePaul University, Chicago, IL 60604, USA. ' Intelligent Multimedia Processing Laboratory, School of Computer Science, Telecommunications, and Information Systems, DePaul University, Chicago, IL 60604, USA. ' Department of Radiology, The University of Chicago, Chicago, IL 60637, USA

Abstract: While the advances in Computed Tomography (CT) technology allow better detection of pulmonary nodules by generating higher-resolution images, the new technology also generates more individual transverse reconstructions. As a result, the efficiency and accuracy of the radiologists interpreting these images is reduced. Double reading by two human observers has been shown to improve the detection of lung cancer. Given the increased cost of double reading and the variation among radiologists| interpretation, the objective is to develop computer-aided tools that could be used as |second readers| when interpreting lung images by apriori rating the nodules based on automatically discovered image-semantics mappings.

Keywords: computed tomography; lung nodules; low-level features; semantic gap; logistic regression; decision trees; SVM; support vector machines; visual ontology; semantics modelling; image data; pulmonary nodules; radiology; lung cancer; lung images; image interpretation; medical imaging.

DOI: 10.1504/IJBET.2010.029653

International Journal of Biomedical Engineering and Technology, 2010 Vol.3 No.1/2, pp.83 - 113

Published online: 30 Nov 2009 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article