Detection of cancerous tumors on chest X-ray images -candidate detection filter and its evaluation

In this paper, we propose new and effective filters for detecting lung nodule candidates on chest X-ray images. They belong to the convergence index filter (CI filter in short). It is a filter that evaluates the degree of convergence of gradient vectors in the neighborhood of the pixel of interest. The degree of convergence is defined by the orientation of the gradient vector. Consequently, the output of this filter does not depend on the contrast of the region, of interest to its background. Three types of CI filter are investigated and their theoretical response characteristics for typical models are shown. Experiments to test their performance have been performed. Their results show the effectiveness of the proposed filters.