Liver cancer detection by using transition features obtained from multi-phase CT images

Abstract

This paper presents a method for automated detection of liver cancer regions based on transition of density at each point obtained from multi-phase X-ray CT images. For describing transition of density, two kinds of feature vectors named Density Transition (DT) and Density Change Transition (DCT) are introduced. DCT is used for extraction of cancer candidates and DT is used for suppression of false candidates. In the experiments using 14 real abdominal CT images with cancer, it was shown that the detection rate was 100% and the number of false-positives was 0.71 regions per case.

Proposal of computer-aided detection system for three dimensional CT images of liver cancer

“…It calculates a convergence index of gradient vectors in the spherical region of Fig. 2 called support region and outputs it to center P. This filter is robust to the changes in the contrast condition and effective in the enhancement of the cancers with low contrast which are difficult to detect by the conventional threshold-based methods [1][2][3]. This process is applied to the two phase images independently and the enhanced images are integrated into one using addition operation which achieved both high sensitivity and low false positive (FP) number in our experiment.…”

“…However it is difficult for physicians to interpret the three dimensional (3D) images obtained with such scanner because of a large number of slice (or section) images. In order to help the physicians in the diagnosis, computer-aided detection (CAD) systems have been proposed [1][2][3][4]. However these systems suffer from several limitations.…”

“…For example, the systems can not be applied to a pair of early and late phase images known to be effective in detecting Hepatocellular Carcinoma (HCC) that is common in Japan. Moreover the threshold-based methods [1][2][3] are sensitive to the contrast condition and it is difficult for them to detect the cancers with low contrast, resulting in low sensitivity.…”