We have recently installed an open access web service to extract chemistry from images and PDF documents, named ChemDocMiner using artificial intelligence. It also extracts small organic molecule chemical structures from the text of suitable PDF documents.
Key Features:
– Document Processing: Upload local images, PDF documents, or fetch patent PDFs directly from the European Patent Office (EPO), USPTO or Lighthouse IP.
– Optical Structure Recognition (OCSR): Uses an ensemble of AI-based tools to identify and extract small molecule chemical structures from visual and textual data.
– Editing and Exporting: Allows users to edit extracted structures and download the results as a CSV file for chemistry research projects.