Linguamatics in Princeton: Unlocking big data’s key insights in life sciences and healthcare

July 27 2015

On July 16, delegates across the life sciences, biotech, healthcare and other knowledge-driven industries gathered in Princeton for Linguamatics’ one-day seminar: “From bench to bedside, unlocking key insights in your data”.  

We heard from Regeneron Pharmaceuticals, Johnson & Johnson, Copyright Clearance Center (CCC) and Linguamatics on how NLP technology is moving into new application areas to improve patient outcomes and unlock key insights across the drug discovery, development and delivery continuum. Delegates were very engaged and many stayed long after the talks had finished, to continue the day’s discussions.  

Jim Dixon, Senior Application Specialist, gave us an introduction to I2E NLP text mining and the new features in the latest I2E release and industry’s first federated text mining platform. Whatever the content, I2E can mine and extract with precision and at scale. You can use Linguamatics I2E to provide valuable intelligence from text, getting you to the answers faster so you can make smarter and better informed decisions.

Dr. Peng Zhang’s presentation showed us a real-life use case of I2E’s potential at Regeneron. Eliminating or modifying a single gene in the mouse genome can provide insight into the role that gene plays in normal physiology and disease pathogenesis, but keeping up-to-date with novel information is time-consuming. Dr. Zhang uses I2E to systematically mine the scientific literature for any reported gene knockout in mice, and associated autoimmune phenotype.

Dr. Zhang said: “Linguamatics I2E exhibited superior performance compared with gold standard MGI resource for identification of knockouts associated with autoimmune phenotypes”

Dr. Peng Zhang presenting at the Linguamatics Princeton seminar

The seminar featured many use cases which showed the return-on-investment delivered by I2E across the drug discovery pipeline and for patient data, but we also were witness to an exciting demonstration of the new RightFind XML for Mining service. This combined CCC-Linguamatics solution enables you to automatically gain text analytics access over full-text content from CCC’s database of millions of articles from major STM publishers.

The last two presentations of the day addressed the increasingly important role of text analytics in tackling Real World Data.

Dr Smita Mitra, Johnson & Johnson, illustrated a valuable use of I2E within J&J Medical Affairs, to pull out key facts around patient-reported outcomes from two sources of textual real world data - voice of the customer information, and narratives from pragmatic clinical trials. She described the major pain points, particularly the volume of text, and the variability and inconsistency of the text. Smita said, "I2E is well on its way to being embedded in our daily workflows/processes" - with increased efficiency of insight generation from these important but otherwise intractable data sources.

Finally, Tony Sheaffer (Senior Director, North American Sales) ran through other customer applications of text analytics over real world data. In particular he discussed the use of NLP at Kaiser Permanente on mining electronic healthcare records for automated identification of patients with specific presentations such as lung nodules; and extracting drug sentiment and effects from social media (Tweets).

The seminar was an excellent warm-up for what will be a ground-breaking Text Mining Summit, October 12-15 2015 in Newport, RI. The conversations will continue with our world-class line up of speakers, roundtable discussions and an inside look at the I2E product roadmap. Delegates will also take text mining into their own hands during the training opportunities available across all experience levels. Some delegates will also be creating a real-time case study for I2E’s ability to address healthcare’s big data challenges during the first-ever I2E Healthcare Hackathon for clinical data.