introduction relationship between climate and health widely studied climatic temperature stress...
TRANSCRIPT
Introduction
• Relationship between climate and health widely studied• Climatic temperature stress increases
cardiovascular disease risk• Solar UV radiation increases skin cancer risk
• Cancer incidence rates expected to rise dramatically
• Cancer risk factors: genetic vs. environmental• Objective: Investigate possible link between
climate and cancer incidence• How: Knowledge discovery via text mining of
biomedical literature• Why: Efficient way to assist scientific
hypothesis formation
Future Work
• Implement investigation linking actual transplant records and cancer incidence data
• Determine if higher association exists between skin cancer and kidney transplants in sunny climates
• Generate new hypotheses that may be tested with traditional, long term epidemiological studies
• Develop model to predict future expected cancer incidence based on environmental factors
1. Crandall, C.G., & Gonzalez-Alonso, J. (2010). Cardiovascular Function in the Heat-Stressed Human. Acta Physiologica, 407-23.
2. Jemal, A., Center, M. M., & DeSantis, C. (2010). Global Patterns of Cancer Incidence and Mortality Rates and Trends. Cancer Epidemiology Biomarkers & Prevention, 1893-907.
3. U.S. cancer statistics: an interactive atlas [Image]. (n.d.). Retrieved from http://apps.nccd.cdc.gov/DCPC_INCA/DCPC_INCA.aspx
4. van der Leun, J. C., & de Gruijl, F. R. (2002). Climate change and skin cancer [Abstract]. Photochemical and Photobiological Sciences, 324-26.
Knowledge Discovery through Text Mining Biomedical Documents Relating Climate and Cancer Incidence
Katherine GasterWofford College
Mentor: Georgia Tourassi, Ph.D.Computational Sciences and Engineering Division
Research Alliance in Math and ScienceOak Ridge National Laboratory
http://info.ornl.gov/sites/rams2012/k_gaster
The Research Alliance in Math and Science program is sponsored by the Office of Advanced Scientific Computing Research, U.S. Department of Energy. The work was performed at the Oak Ridge National Laboratory, which is managed by UT-Battelle, LLC under Contract No. De-AC05-00OR22725. This work has been authored by a contractor of the U.S. Government, accordingly, the U.S. Government retains a nonexclusive, royalty-free license to publish or reproduce the published form of this contribution, or allow others to do so, for U.S. Government purposes. The author would like to thank Dr. Georgia Tourassi and Dr. Robert Patton for their support and assistance with this research. I would also like to thank Kara Kruse and Rashida Askia for their continued support and assistance, and Debbie McCoy for the opportunity to perform this research and her continued support and assistance.
MethodologyMethodology
Study design• Literature search to find relevant publications• Text mining using Piranha software • Investigate impact of Piranha’s key features
Changingthreshold values
Changing stopword groups
Categoryaddition
• Values from 0.0 to 1.0; intervals of 0.2
• Intervals halved three times
• Combinations of groups• In conjunction with varied
threshold values• Cancer types; risk factors• Used with varied
threshold values; stop word groups
Figure 2. Expected association: breast or prostate cancer risk and genetic profile.
Total documents: 6Total Piranha Categories: 1unknown 100%Total Document Categories: 1All Article… 100%Top Terms:herit, heritable,prostate, pros …genetics, ge ne…brcapopulation-based
Results
Discovered expected associations• Skin cancer risk and UV exposure• Breast cancer / prostate cancer risk and genetic
profile
Total documents: 3Total Piranha Categories: 1Unknown 100%Total Document Categories: 1All Article… 100%Top Terms:uvphotobiol, phot….rscarcinogenesisbl ish ed
Total documents: 3Total Piranha Categories: 1unknown 100%Total Document Categories: 1All Article… 100%Top Terms:renalnmsc, nmsc:, nmscsdermatolkeratoses:, ker…transplanted, t…
Melanoma of the Skin Incidence Rates by State, 2008
Figure 1. Expected association: cancer risk and UV exposure.
Figure 4. Unexpected association: kidney transplants and skin cancer.
Figure 3. Map depicting relative melanoma incidence for each U.S. state in 2008.
Discovered unexpected association: increased skin cancer risk in kidney transplant recipients
• Obtained additional publications to investigate association further• Completed literature search for publications
referring to kidney transplants and skin cancer• Analyzed supplementary documents with and
without original documents
Conclusions
• Text mining revealed expected associations between climatic factors and cancer types
• Unexpected association discovered: increased skin cancer risk in kidney transplant recipients
• Supplementary articles clustered as expected
Figure 6. Document clustering of original and supplementary articles. Original threshold values, no stop words.
Figure 5. Document clustering of original articles. Original threshold values, no stop words.