NLM in Focus is pleased to share this profile of Dr. Zhiyong Lu, PhD, senior investigator at NLM’s National Center for Biotechnology Information. For more information about Dr. Lu and his team, check out this Quick Q&A.

Education: Nanjing University, Nanjing, China (BS in computer science); University of Alberta, Edmonton, Alberta, Canada (MS in computer science); University of Colorado School of Medicine, Aurora, Colorado (PhD in bioinformatics)

Came to NIH: In 2007 as a staff scientist; became an associate investigator in 2009; became a Stadtman investigator in 2011

Selected professional activities: Associate editor, BMC Bioinformatics and Journal of Healthcare Informatics Research; Editorial board member, Database; Steering committee member, BioCreative; PI, PubMed Labs

Outside interests: Playing with his three young children; swimming

Website: https://irp.nih.gov/pi/zhiyong-lu; https://www.ncbi.nlm.nih.gov/bionlp/

Research interests: I am directing text-mining research and lead the new overall efforts to improve and rebuild PubMed. (Our current development is an innovative system called PubMed Labs in which we are experimenting with new ways to improve search quality and usability for biomedical literature.) Text mining involves going through digitized and unstructured text to find useful, high-quality information. Text analysis includes retrieving relevant documents; identifying named entities (such as gene and disease names); and extracting relationships between entities and other natural language processing tasks. The technology is now applied to a broad range of government, research, and business and marketing needs, and it is used in a wide variety of real-world applications (such as automatic spam filtering or predicting flu trends). In the health-care industry, text mining is essential in helping to find important information that’s buried within tens of millions of biomedical articles.

My research group is developing new computational methods and software tools to analyze and make sense of free text data in scholarly publications and other biomedical texts such as electronic medical records. We apply text-mining research to improve biomedical literature search; assist in the manual curation of biological databases; and predict new uses of existing drugs. Our text-mining software tools and web services, such as PubTator, have been widely used (over 100 million requests since 2015) by scientists from around the world and have also been integrated into PubMed, PubChem, and many other NCBI web resources.

I also co-organize international scientific conferences and community-wide challenges such as BioCreative, the longest-running international event for evaluating text-mining and information-extraction systems applied to the biomedical domain.

This article first appeared in the NIH Catalyst, May-June 2017 issue. It is reposted with permission. Dr. Lu was also included in a story on the Stadtman Investigators in the January-February 2013 issue of NIH Catalyst.