Lung Cancer Detection from X-Ray CT Images using Pixel-based Support Vector Machine

H. Takizawa, H. Nishizako, S. Wada, and T. Matsumoto (Japan)

Keywords

Thoracic CT images, Lung cancer, Discriminant filters, Pixel-based support vector machine

Abstract

This paper describes a detection method of pulmonary nod ules (i.e. possible cancers) from chest X-ray computed tomography (CT) images. The method consists of two phases. First, initial nodule candidates are detected by use of circle- and ring-shaped discriminant filters. Next, false positives such as blood vessels are reduced by use of radially-arranged rectangular discriminant filters and a pixel-based support vector machine. The proposed method is applied to actual CT images, and experimental results are shown.

Important Links:

• DOI: 10.2316/P.2010.710-048
• From Proceeding (710) Signal and Image Processing - 2010

Go Back