Evaluation of Multimodal Algorithms for the Segmentation of Multiparametric MRI Prostate Images.

Evaluation of Multimodal Algorithms for the Segmentation of Multiparametric MRI Prostate Images. Comput Math Methods Med. 2020;2020:8861035 Authors: Nai YH, Teo BW, Tan NL, Chua KYW, Wong CK, O'Doherty S, Stephenson MC, Schaefferkoetter J, Thian YL, Chiong E, Reilhac A Abstract Prostate segmentation in multiparametric magnetic resonance imaging (mpMRI) can help to support prostate cancer diagnosis and therapy treatment. However, manual segmentation of the prostate is subjective and time-consuming. Many deep learning monomodal networks have been developed for automatic whole prostate segmentation from T2-weighted MR images. We aimed to investigate the added value of multimodal networks in segmenting the prostate into the peripheral zone (PZ) and central gland (CG). We optimized and evaluated monomodal DenseVNet, multimodal ScaleNet, and monomodal and multimodal HighRes3DNet, which yielded dice score coefficients (DSC) of 0.875, 0.848, 0.858, and 0.890 in WG, respectively. Multimodal HighRes3DNet and ScaleNet yielded higher DSC with statistical differences in PZ and CG only compared to monomodal DenseVNet, indicating that multimodal networks added value by generating better segmentation between PZ and CG regions but did not improve the WG segmentation. No significant difference was observed in the apex and base of WG segmentation between monomodal and multimodal networks, indicating that the segmentations at the apex and base were more...
Source: Computational and Mathematical Methods in Medicine - Category: Statistics Tags: Comput Math Methods Med Source Type: research