Texture-Based Classification of Hepatic Primary Tumors in Multiphase CT

A new approach to the hepatic primary tumor recognition from dynamic CT images is proposed. In the first step of the proposed method, texture features are extracted from manually traced ROI-s to objectively characterize lesions. The second step consists in applying decision tree classifier. For the first time, the parameters obtained in subsequent acquisition moments are analyzed simultaneously.

• Type: Conference
• Year: 2004
• Where: MICCAI
• Authors: Dorota Duda, Marek Kretowski, Johanne Bézy-Wendling
• Keywords: Decision Tree Classifier, Hepatic Primary Tumor, Medical Imaging, MICCAI 2004, Subsequent Acquisition Moments