2020/05/15 by Kelei He, He, Kelei, Chunfeng Lian +13
Computer Science · Dentistry · Engineering · Medicine · #Computer Vision and Pattern Recognition (cs.CV) #Dental Radiography and Imaging #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Machine Learning (cs.LG) #Medical Imaging and Analysis #Prostate Cancer Diagnosis and Treatment #cs.CV #cs.LG #eess.IV #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.2005.07462
openalex publication_date 2020/05/15 · openalex created_date 2020/07/16 · arxiv created 2021/01/23 · arxiv updated 2021/01/26 · openalex updated_date 2026/07/28
Fully convolutional networks (FCNs), including UNet and VNet, are widely-used network architectures for semantic segmentation in recent studies. However, conventional FCN is typically trained by the cross-entropy or Dice loss, which only calculates the error between predictions and ground-truth labels for pixels individually. This often results in non-smooth neighborhoods in the predicted segmentation. To address this problem, we propose a two-stage framework, with the first stage to quickly localize the prostate region and the second stage to precisely segment the prostate by a multi-task UNet architecture. We introduce a novel online metric learning module through voxel-wise sampling in the multi-task network. Therefore, the proposed network has a dual-branch architecture that tackles two tasks: 1) a segmentation sub-network aiming to generate the prostate segmentation, and 2) a voxel-metric learning sub-network aiming to improve the quality of the learned feature space supervised by a metric loss. Specifically, the voxel-metric learning sub-network samples tuples (including triplets and pairs) in voxel-level through the intermediate feature maps. Unlike conventional deep metric learning methods that generate triplets or pairs in image-level before the training phase, our proposed voxel-wise tuples are sampled in an online manner and operated in an end-to-end fashion via multi-task learning. To evaluate the proposed method, we implement extensive experiments on a real CT image dataset consisting of 339 patients. The ablation studies show that our method can effectively learn more representative voxel-level features compared with the conventional learning methods with cross-entropy or Dice loss. And the comparisons show that the proposed method outperforms the state-of-the-art methods by a reasonable margin.