Liver tumor detection and segmentation using kernel-based extreme learning machine

Abstract

In this project, a semi-automatic approach of the detection and segmentation of liver tumors from 3D computed tomography (CT) images is presented. The automatic detection of liver tumor can be formulized as a novelty detection or two-class classification issue. The method can also be used for tumor segmentation, where each voxel is to be assigned with a correct label, either a tumor class or a non-tumor class. A voxel is represented with a rich feature vector that distinguishes itself from voxels in different classes. A fast learning algorithm Extreme Learning Machine (ELM) is trained as a voxel classifier. In automatic liver tumor detection, we propose and show that ELM can be trained as a one-class classifier with only healthy liver samples in the training dataset. It results in a method of tumor detection based on novelty detection. Then we compare it with the two-class ELM detection case. To extract the boundary of a tumor, we adopt the semi-automatic approach by randomly selecting samples in 3D space within a limited region of interest (ROI) for classifier training. Our approach is validated on a group of patients’ CT data and the experiment shows good detection and encouraging segmentation results. Part of the work presented in this FYP report was accepted as a conference paper [23] to be presented at the 35th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC'13) to be held in Osaka International Convention Center, in Osaka, Japan on July 3-7, 2013 [22].