A new NMF-autoencoder based CAD system for early diagnosis of prostate cancer

In this paper, we propose a novel non-invasive framework for the early diagnosis of prostate cancer from diffusion-weighted magnetic reasoning imaging (DW-MRI). The proposed approach consists of three main steps. In the first step, the prostate is localized and segmented based on a new level-set model. This model is guided by a stochastic speed function that is derived using nonnegative matrix factorization (NMF). The NMF attributes are calculated using information from the MRI intensity, a probabilistic shape model, and the spatial interactions between prostate voxels. In the second step, the apparent diffusion coefficient (ADC) of the segmented prostate volume is mathematically calculated for different b-values. To preserve continuity, the calculated ADC values are normalized and refined using a Generalized Gauss-Markov Random Field (GGMRF) image model. The cumulative distribution function (CDF) of refined ADC for the prostate tissues at different b-values are then constructed. These CDFs are considered as global features which can be used to distinguish between benign and malignant tumors. Finally, a deep learning auto-encoder network, trained by a non-negativity constraint algorithm (NCAE), is used to classify the prostate tumor as benign or malignant based on the CDFs extracted from the previous step. Preliminary experiments on 42 clinical DW-MRI data sets resulted in 97.6% correct classification (sensitivity = 100% and specificity = 95.24%), indicating the high accuracy of the proposed framework.