In vivo MRI based prostate cancer identification with random forests and auto-context model

Abstract

Prostate cancer is one of the major causes of cancer death for men. Magnetic Resonance (MR) image is being increasingly used as an important modality to detect prostate cancer. Therefore, identifying prostate cancer in MRI with automated detection methods has become an active area of research, and many methods have been proposed to identify the prostate cancer. However, most of previous methods only focused on identifying cancer in the peripheral zone, or classifying suspicious cancer ROIs into benign tissue and cancer tissue. In this paper, we propose a novel learning-based multi-source integration framework to directly identify the prostate cancer regions from in vivo MRI. We employ the random forest technique to effectively integrate features from multi-source images together for cancer segmentation. Here, the multi-source images include initially only the multi-parametric MRIs (T2, DWI, eADC and dADC) and later also the iteratively estimated and refined tissue probability map of prostate cancer. Experimental results on 26 real patient data show that our method can accurately identify the cancerous tissue.