Nhu Le

Distinguished Scientist

Research Roles:

• Distinguished Scientist, Cancer Control Research, BCCRC
• Adjunct Professor, Statistics, UBC
   

Education:

• PhD (Statistics), University of Washington, 1990
• MSc (Statistics), University of British Columbia, 1986
• BSc (Computer Science & Mathematics), University of British Columbia, 1984

Research Interests:

•  Biostatistics & Statistical Genetics
• Gene-environment interactions in Cancer
• Identification of Occupational and Environmental Cancer Risk Factors
• Air Pollution Exposure Assessment &Health Impact of Air Pollution
• Detection of Spatial and Temporal Clustering

My current research interests are primarily in the areas of biostatistics, statistical genetics, time series and spatial statistics with emphasis on application to occupational and environmental cancer epidemiology, including gene-environment interactions. The work motivated by occupational oncology studies carried out in the Cancer Control Research Program concerns with the possibility of characterizing occupational cancer risk factors and identifying potential carcinogens in the workplace relevant to specific industrial context of BC. The overall objective is the reduction of risk.

Issues arising from environmental studies have led to the development of a theory for spatial interpolation. The work allows for the development of spatial prediction distributions for regions with no monitoring stations using data available at only a few monitoring stations. This theory is particularly useful in environmental cancer epidemiology where pollution data are available only at a few locations. Another consequence of this work is the development of methods for redesigning environmental monitoring network where one may want to with extend or contract an existing network where one may want to either extend or contract an existing network. Various theoretical extensions of the theory and their potential applications are being explored. In particular, the methodology will be used in the on-going case-control study to identify environmental risk factors and gene-environment interactions in cancer.

Other research interests include the use of mixture distribution in modeling overdisperse Poisson counts or non-Gaussian time series, and the use of robust Bayes factors in model comparison for autoregressive processes. The research has led to the development of a cancer cluster detection methodology, which allows for administrative regions to have substantially different sizes and populations such as those in BC.