DECIMER: towards deep learning for chemical image recognition
- PDF / 6,105,637 Bytes
- 9 Pages / 595.276 x 790.866 pts Page_size
- 35 Downloads / 208 Views
(2020) 12:65 Rajan et al. J Cheminform https://doi.org/10.1186/s13321-020-00469-w
Open Access
PRELIMINARY COMMUNICATION
DECIMER: towards deep learning for chemical image recognition Kohulan Rajan1, Achim Zielesny2 and Christoph Steinbeck1*
Abstract 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 Deep lEarning for Chemical ImagE Recognition (DECIMER), 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. It translates a bitmap image of a molecule, as found in publications, into a SMILES. 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 superior over SMILES and we have a preliminary indication that the recently reported SELFIES outperform DeepSMILES. An extrapolation of our results towards larger training data sizes suggests that we might be able to achieve near-accurate prediction with 50 to 100 million training structures. This work is entirely based on open-source software and open data and is available to the general public for any purpose. Keywords: Optical chemical entity recognition, Chemical structure, Deep learning, Deep neural networks, Autoencoder/decoder Main text The automatic recognition of chemical structure diagrams from the chemical literature (herein termed Optical Chemical Entity Recognition, OCER) is an indispensable component of workflows to re-discover information about chemicals and to make it available in open-access databases. While the chemical structure is often at the heart of the findings reported in chemical articles, further information about the structure is present either in textual form or in other types of diagrams such as titration curves, spectra, etc. (Fig. 1). Previous software systems for OCER have been described and were both incorporated into commercial and open-source systems. These software systems *Correspondence: christoph.steinbeck@uni‑jena.de 1 Institute for Inorganic and Analytical Chemistry, Friedrich-SchillerUniversity Jena, Lessingstr. 8, 07743 Jena, Germany Full list of author information is available at the end of the article
include Kekulé [1, 2], the Contreras system [3], the IBM system [4], CLIDE [5] as well as the open-source approaches chemOCR [6–8], ChemReader [9], OSRA [10] and ChemRobot described in a patent [11]. All of these software packages share a general approach to the problem, comprising the steps (a) scanning, (b) vectorization, (c) searching for dashed lines and dashed wedges, (d) character recognition, (e) graph compilation, (f ) post-processing, (g) display and editing.
Data Loading...
