Improved Minimum-Volume Enclosing Simplex Algorithm for Prostate DCE-MRI Analysis

The purpose of this study is to explore the potential of dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) for classification of prostate cancer using an improved minimum-volume enclosing simplex (MVES). A total of 15 DCE sequences were collected from 5 prostate cancer (PCa) patients and 10 non-PCa patients in this retrospective study. After DCE signal intensity within the prostate was converted to contrast agent concentration, an improved MVES algorithm with a noise regular term was implemented to decompose concentration curves of the prostate into 3 components. Taking one of the component curves as the arterial input function, the other two curves were fitted to the Tofts and Kermode model and pharmacokinetic parameters K
^trans
and k
_ep
were obtained. Statistical analysis indicated that for the component curves, K
^trans
and k
_ep
were significantly different between the prostates with and without cancer, with p-values of 0.003 and 0.015, respectively. We conclude that the improved MVES has potential for separating prostates with and without cancer.