文献类型：期刊文献

英文题名：SwinOCSR: end-to-end optical chemical structure recognition using a Swin Transformer

作者：Xu, Zhanpeng[1];Li, Jianhua[1];Yang, Zhaopeng[1];Li, Shiliang[2];Li, Honglin[2]

机构：[1]East China Univ Sci & Technol, Sch Informat Sci & Engn, 130 Mei Long Rd, Shanghai 200237, Peoples R China;[2]East China Univ Sci & Technol, Sch Pharm, State Key Lab Bioreactor Engn, Shanghai Key Lab New Drug Design, Shanghai 200237, Peoples R China

年份：2022

卷号：14

期号：1

外文期刊名：JOURNAL OF CHEMINFORMATICS

收录：;WOS:【SCI-EXPANDED(收录号:WOS:000819781000001)】；

基金：National Key R&D Program of China (under Grant No. 2016YFA0502304) and Important Drug Development Fund, Ministry of Science and Technology of China (2018ZX09735002).

语种：英文

外文关键词：Chemical Structure Recognition; Deep Learning; Swin Transfromer; End-to-End Model

摘要：Optical chemical structure recognition from scientific publications is essential for rediscovering a chemical structure. It is an extremely challenging problem, and current rule-based and deep-learning methods cannot achieve satisfactory recognition rates. Herein, we propose SwinOCSR, an end-to-end model based on a Swin Transformer. This model uses the Swin Transformer as the backbone to extract image features and introduces Transformer models to convert chemical information from publications into DeepSMILES. A novel chemical structure dataset was constructed to train and verify our method. Our proposed Swin Transformer-based model was extensively tested against the backbone of existing publicly available deep learning methods. The experimental results show that our model significantly outperforms the compared methods, demonstrating the model's effectiveness. Moreover, we used a focal loss to address the token imbalance problem in the text representation of the chemical structure diagram, and our model achieved an accuracy of 98.58%.