A new analytical tool from IBM is helping scientists scan patents and other intellectual property for information about molecular data. The cloud-based strategic IP insight platform, or SIIP, uses image analysis and enhanced optical recognition of chemical images and symbols to quickly obtain the information, a process that otherwise could take months to do manually.
The platform, housed in the IBM SmartCloud, processes documents in the U.S. Patent and Trademark Office, the European Patent Office and the World Intellectual Property Organization, as well as the public domain documents in the U.S. National Library of Medicine's MEDLINE database.
Originally In-House Tool
SIIP was originally developed for IBM'S own use, and it has been used by hundreds of IP experts in the company for management and licensing. The company said that many top life-sciences companies also use SIIP because of its special focus on sifting patents for biological and chemical information.
SIIP can process molecular structures drawn as chemical compounds, and is being used to extract molecular and chemical information from patents, abstracts and other documents within hours of their publication. To date, more than 200 million chemical instances, gleaned from more than 30 million documents, have been extracted.
A user can perform analytics through a Web browser. Searches can be conducted via a drug's trade name, generic name, chemical ID number, or even by drawing the chemical structure inside the tool. Identical or similar compounds will be returned within seconds. The tool can also show trends in the industry, a competitive landscape, a prior art search and other analysis.
Via the IBM SmartCloud, the information can be shared within an organization, or among organizations. The company said that the tool improves R&D productivity, increases IP licensing, supports open innovation initiatives, helps to identify potential collaborators or acquisition targets, presents competitive intelligence, and increases the rate at which patents and grants are awarded.
Contribution to NIH
Because of the value of the retrieved information to the scientific community, IBM has donated to the National Institutes of Health an SIIP-created database of more than 2.5 million unique chemical compounds, along with the references in the literature. The database, extracted from million of patents and scientific documents, is expected to allow medical researchers to more readily understand relationships between chemical compounds.
Dr. Ying Chen, IBM research scientist, said the database donated to NIH will be used in conjunction with NIH's existing "rich set of content," such as comparing uses for a given compound across the literature.
Chen said another major differentiator for SIIP is the ability for natural language processing, instead of the keyword-based searching common in other such search engines. She also noted that SIIP has applications in other industries that use chemical information, such as consumer electronics, automotive, or oil and gas companies, and that the software-as-a-service tool "can be extremely useful for smaller companies as well as larger ones."