Predictive Analytics Helps Spur Cancer ResearchPredictive Analytics Helps Spur Cancer Research

Children's Memorial Research Center is leveraging SPSS predictive analytics to differentiate pediatric brain tumor types. By analyzing both patient data and medical research literature with SPSS Inc. technologies, Children's Memorial Research Center has gained unique insights into tumor classification and treatment strategies.

Children's Memorial Research Center is the research arm of Children's Memorial Hospital, and the center for pediatric research at Northwestern University.

Last month, Dr. Eric Bremer, director of the Brain Tumor Research Program at Children's Memorial Research Center, presented the results of his SPSS-based research at United Devices' 2005 Life Sciences Grid Conference in Zurich, Switzerland.

Bremer uses SPSS data mining technology, Clementine, to analyze and classify pediatric brain tumor types. Additionally, he employs SPSS text mining software, LexiQuest Mine, to discover previously overlooked relationships contained in volumes of biomedical literature.

Bremer required a methodology for conducting faster analysis of research literature spanning a five-year period. He selected United Devices, a provider of enterprise grid solutions, to grid-enable LexiQuest Mine on the company's Grid MP solution. This also supports Bremer's long-term goal of conducting continuous, comprehensive textual analysis on a larger number of medical journals.

Bremer cited an example from a pilot project where he was able to process 124,000 medical abstracts in exactly one hour and 18 minutes. He contrasted this grid-enabled analysis with the average time it would take him to conduct identical analysis on a stand-alone computer, between 20 and 24 hours.

"The automated extraction of information from biomedical literature promises to play an increasingly important role in text-based knowledge discovery processes," reported Bremer, in a press release. "Leveraging SPSS predictive analytics and United Devices' expertise in grid computing, we've developed an integrated technology system that can efficiently extract and organize gene relationships from full text articles. We can also correlate this insight with both past and ongoing research on effective pediatric cancer treatments."