An application of cascaded 3D fully convolutional networks for medical image segmentation

14 Mar 2018Holger R. RothHirohisa OdaXiangrong ZhouNatsuki ShimizuYing YangYuichiro HayashiMasahiro OdaMichitaka FujiwaraKazunari MisawaKensaku Mori

Recent advances in 3D fully convolutional networks (FCN) have made it feasible to produce dense voxel-wise predictions of volumetric images. In this work, we show that a multi-class 3D FCN trained on manually labeled CT scans of several anatomical structures (ranging from the large organs to thin vessels) can achieve competitive segmentation results, while avoiding the need for handcrafting features or training class-specific models... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
3D Medical Imaging Segmentation TCIA Pancreas-CT Multi-class 3D FCN Dice Score 76.8 # 2