FSU team wins global competition to develop next-gen tools, improve public health solutions – Florida State News

A Florida State University team has won an international competition for developing software tools that generate visual aids demonstrating the complex connections among vast amounts of biomedical research.

The team of 12, led by FSU Professor of Statistics Jinfeng Zhang, took home first place in the LitCoin Natural Language Processing (NLP) Challenge, organized by the National Institutes of Health’s National Center for Advancing Translational Sciences (NCATS) in partnership with the NASA Tournament Lab.

The NIH and NASA invited teams to develop software solutions that can connect biomedical concepts from various collections of open data repositories with the goal of improving public health solutions. Scientists are dealing with more data than ever before, but it has also become a monumental task to sift through it. Through this process, teams tackled ways to make medical research easier to navigate and identify.

FSU competed against teams from across the country and around the world, including Australia, the Netherlands and Portugal.

“We are very happy to win but the most important thing is that we learned a lot from this, Zhang said. “I always tell my students that. It’s great to come in first place but I think the experience, the knowledge we learned, that is what’s most valuable.”

Xufeng Niu, chair of FSU’s Department of Statistics, said the challenge aims to tackle inherent problems with the way data is managed and accessed by researchers, specifically in the vast realm of biomedical research, where such assets are so valuable to tackling public health issues.

“Understanding a large amount of biomedical data from numerous medical studies is scientifically essential and statistically challenging,” Niu said. “Dr. Jinfeng Zhang and his team have been working on big data analysis, text mining, and search engine optimization for the last 10 years.”

The FSU team also included several students and a handful of faculty members and alumni who wanted to get involved after hearing about the challenge. Team members included:

  • Jinfeng Zhang, professor, Department of Statistics
  • Yuan Zhang, research faculty, Department of Statistics
  • Feng Pan, research scientist, Department of Statistics
  • Xin Sui, graduate student, Department of Statistics
  • Keqiao Li, graduate student, Department of Statistics
  • Shubo Tian, graduate student, Department of Statistics
  • Arslan Erdengasileng, graduate student, Department of Statistics
  • Qing Han, graduate student, Department of Statistics
  • Wanjing Wang, graduate student, Department of Statistics
  • Kaixian Yu, alumnus, Department of Statistics
  • Jian Wang, alumnus, Department of Mathematics
  • Tingting Zhao, associate professor, Department of Geography

The competitors used information from published abstracts to create natural language processing systems that could recognize biomedical concepts and the relationships between them.

NCATS said the software packages built for the NLP Challenge will be used to help create web development tools with the goal of creating a new type of biomedical publication. This will encourage sharing of computationally accessible data from the time of publication, with the aim of changing the way research results are compiled for use when developing public health policies and new treatments and cures.

Niu said the team’s winning work exemplifies how the department is making a direct impact on the international research community.

“Dr. Zhang’s team winning the LitCoin Natural Language Processing challenge is a huge breakthrough. My department and I are very proud of this significant achievement,” Niu said. “We expect Dr. Zhang’s team to continue its outstanding performance in developing cutting-edge techniques in the field of biomedical data analysis and new medical treatment identification.”

For more information on the NLP Challenge, and to view the complete list of winning teams, visit ncats.nih.gov.

Leave a Reply

Your email address will not be published.