DECIMER - Towards Deep Learning for Chemical Image Recognition

The automatic recognition of chemical structure diagrams from the literature is an indispensable component of workflows to re-discover information about chemicals and to make it available in open-access databases. Here we report preliminary findings in our development of DECIMER (Deep lEarning for Chemical ImagE Recognition), a deep learning method based on existing show-and-tell deep neural networks which makes very few assumptions about the structure of the underlying problem. The training state reported here does not yet rival the performance of existing traditional approaches, but we present evidence that our method will reach a comparable detection power with sufficient training time. Training success of DECIMER depends on the input data representation: DeepSMILES are clearly superior over SMILES and we have preliminary indication that the recently reported SELFIES outperform DeepSMILES. An extrapolation of our results towards larger training data sizes suggest that we might be able to achieve >90% accuracy with about 60 to 100 million training structures, so that training can be completed within several months on a single GPU. This work is completely based on open-source software and open data and is available to the general public for any purpose.