Surgical instrument segmentation based on multi-scale and multi-level feature network

Yiming Wang, Zhongxi Qiu, Yan Hu, Hao Chen, Fangfu Ye, Jiang Liu

Abstract: Surgical instrument segmentation is critical for the field of computer-aided surgery system. Most of deep-learning based algorithms only use either multi-scale information or multi-level information, which may lead to ambiguity of semantic information. In this paper, we propose a new neural network, which extracts both multi-scale and multilevel features based on the backbone of U-net. Specifically, the cascaded and double convolutional feature pyramid is input into the U-net. Then we propose a DFP (short for Dilation Feature-Pyramid) module for decoder which extracts multi-scale and multi-level information. The proposed algorithm is evaluated on two publicly available datasets, and extensive experiments prove that the five evaluation metrics by our algorithm are superior than other comparing methods.