Automated periodontitis bone loss diagnosis in panoramic radiographs using a bespoke two-stage detector

Zhengmin Kong, Hui Ouyang, Yiyuan Cao, Tao Huang, Euijoon Ahn, Maoqi Zhang, Huan Liu

Abstract: Highlights•A novel detector is proposed for periodontitis detection (PDCNN).•PDCNN is combined with a proposal-connection module to perform better.•We construct a new dental-image dataset with periodontitis annotations.•Experiments show that PDCNN outperforms state-of-the-art detectors.•We made this dataset and our code publicly available on GitHub.