Model-free prostate cancer segmentation from dynamic contrast-enhanced MRI with recurrent convolutional networks: A feasibility study. (July 2019)

• Record Type: Journal Article
• Authors: Lee, Peter Q.; Guida, Alessandro; Patterson, Steve; Trappenberg, Thomas; Bowen, Chris; Beyea, Steven D.; Merrimen, Jennifer; Wang, Cheng; Clarke, Sharon E.
• Is Part Of: Computerized medical imaging and graphics. Volume 75(2019)
• Page Start: 14
• Page End: 23
• Publication Date: 2019-07
• Subjects: Magnetic resonance imaging -- Dynamic contrast enhancement -- Prostate cancer -- Recurrent convolutional networks; Diagnostic imaging -- Periodicals; Imaging systems in medicine -- Periodicals; Diagnosis, Radioscopic -- Data processing -- Periodicals; Diagnostic Imaging -- Periodicals; Imagerie pour le diagnostic -- Périodiques; Diagnostic imaging; Periodicals; Electronic journals; Electronic journals; 616.0754
• DOI: 10.1016/j.compmedimag.2019.04.006
• Languages: English
• ISSNs: 0895-6111

Abstract

Highlights: Recurrent fully convolutional network used to segment prostate cancer using DCE-MRI time series. Proposed recurrent convolutional network using DCE statistically outperforms segmentation of a fully convolutional network using Ktrans. Suggests that Tofts model loses useful information in raw DCE time series. Abstract: Dynamic contrast enhanced (DCE) magnetic resonance imaging (MRI) is a method of temporal imaging that is commonly used to aid in prostate cancer (PCa) diagnosis and staging. Typically, machine learning models designed for the segmentation and detection of PCa will use an engineered scalar image called K trans to summarize the information in the DCE time-series images. This work proposes a new model that amalgamates the U-net and the convGRU neural network architectures for the purpose of interpreting DCE time-series in a temporal and spatial basis for segmenting PCa in MR images. Ultimately, experiments show that the proposed model using the DCE time-series images can outperform a baseline U-net segmentation model using K trans . However, when other types of scalar MR images are considered by the models, no significant advantage is observed for the proposed model.