Cross-device automated prostate cancer localization with multiparametric MRI

Automated cancer localization with supervised techniques plays an important role in guiding biopsy, surgery and treatment. It is crucial to have an accurate training dataset for supervised techniques. Since different devices with e.g. different protocols and/or field strengths cause different intensity profiles, each device/protocol must have an accompanying training dataset which is very costly to obtain. In this paper, we propose a novel method that has the ability to design classifiers obtained from one imaging protocol and/or MRI device to be used on a dataset from another protocol and/or imaging device. As an example problem we consider prostate cancer localization with multiparametric MRI. We show that simple normalization techniques such as z-score are not sufficient to allow for cross-device automated cancer localization. On the other hand, the methods we have originally developed based on relative intensity allows us to successfully use a classifier obtained from one device to be applied on a test patient imaged with another device.