Medical object detector jointly driven by knowledge and data

Xianhua Zeng, Yuhang Liu, Jian Zhang, Yongli Guo

Abstract: Highlights•Proposed a medical object detector jointly driven by knowledge and data.•Designed an effective mechanism to represent and fuse external semantic information.•Designed a fusion module to fuse professional medical knowledge with data.•A loss for unbalanced data that enables a reasonable fusion of knowledge and data.•Experiments show that our model outperforms other models in medical scenarios.