At the Center for Translational Research of the Tokyo University Institute of Medical Science, I am responsible for research design and statistical analysis in clinical trials and epidemiological studies. I have been published papers through collaborative research. My main focus on the statistical supporting is discerning true correlations by taking into account confounding factors and correlations between variables, and constructing statistical models that are beneficial to medicine and society. I am also engaged in environmental epidemiology and research using omics data such as metabolomics. Beyond basic biostatistical methods, we incorporate data mining techniques like clustering and machine learning methods like LASSO and random forest in our analysis. Additionally, I have a long-standing involvement in cancer epigenetics, particularly in the study of DNA methylation. Our recent studies integrate microarray data of DNA methylation with gene expression and miRNA (microRNA) expression data to statistically model gene expression control in cancer tissues (Nojima et.al, Mol Cancer 2016). Lately, I've been focusing on database epidemiology using Japan's Ministry of Health, Labour and Welfare NDB database (https://www.mhlw.go.jp/stf/seisakunitsuite/bunya/kenkou_iryou/iryouhoken/reseputo/index.html) to examine the effectiveness of specific health check-ups and guidance. As database epidemiology research is still underdeveloped in Japan, which we believe has impacted the lack of impactful clinical research during the COVID-19 crisis. The research topics I work on are diverse, but all of them have data science as the central methodology. I generally use big data for statistical analysis on topics of interest. With a background in internal medicine, I am keep focusing on health science by using various measurements with clinical information. I welcome individuals who wish to learn data science to answer research questions that are relevant as a healthcare provider and epidemiologist. For example, research topics may include approaches that explore the relationship between specific characteristics of omics data and clinical outcomes, as well as medical phenotypes.