Early Diagnosis of Hepatocellular Carcinoma Using Machine Learning Method

Liver cancer is the fourth most common cause of death from cancer and over 90% of cases are hepatocellular carcinoma (HCC). The diagnosis of HCC at an early stage is very important to improve live cancer treatment. Thus, it is highly desirable to develop a signature to aid the early diagnosis of HCC under these clinical scenarios. Several conventional methods have been used for discriminating HCC from cirrhosis tissues of non-HCC patients (CoHCC). However, the recognition successful rates are still far from satisfactory. In this study, we applied a computational approach that combined support vector machine (SVM) with maximum relevance minimum redundancy (mRMR) algorithm and incremental feature selection (IFS) technique to a set of microarray data generated from 1091 HCC samples and 242 CoHCC samples. The within-sample relative expression orderings (REOs) method was used to extract numerical descriptors from gene expression profiles. After removing the unrelated features by using mRMR with IFS, we achieved "11-gene-pair" which could produce outstanding results. We further investigated the discriminate capability of the "11-gene-pair" for HCC recognition on several independent datasets. The wonderful results were obtained, demonstrating that the selected gene pairs can be signature for HCC. It is anticipated that the proposed computational model might be practical and effective for aiding the early diagnosis of HCC in multiple datasets of both biopsy and surgically resected samples, which is also suitable for minimum biopsy specimens and inaccurately sampled specimens.