1. HOME
  2. Development and application of a disease-clustering statistical method using scan statistics

Development and application of a disease-clustering statistical method using scan statistics

  • School of Medicine/Graduate School of Medicine
  • Program in Integrated Medicine
  • Clinical Pharmacology
  • (Biostatistics)

Shigeyuki Matsui [Professor]

http://www.nagoya-biostat.jp/english/index.html

Outline of Seeds

In spatial epidemiology research, regions of significantly higher (or lower) mortality than other regions are identified to examine disease clustering. Detection of disease clustering is important for symptom surveillance research in order to prevent infection or bioterrorism. We are developing flexible scan statistics using scan statistics and exploring its application.

Novelty and originality of this research

In this study we developed a method for more precise detection of disease clustering. We also are proposing an assessment method and developing analysis software. The developed software is used in epidemiological studies and symptom surveillance both in Japan and beyond.

Application and research area for Industry collaboration

We developed methodology and software and are exploring application thereof in various fields. For example, surveying actual regional characteristics of diseases including intractable diseases, detection of disease related-genes, brain image analysis and signal detection based on spontaneous report of side effects brought on by pharmaceutical products were performed in our study.

Key Takeaway

Prediction of biological phenomenon in humans is an extremely difficult leading to significant uncertainty due to the complexity of mechanisms and significant individual differences. However, even though these issues are true, when we actually collect and analyze biological phenomenon data, we can scientifically approximate certain kinds of principles present in the background of the phenomenon. Biostatistics provides the theory and effective methodology in this process to enable scientific inference and prediction, taking account of ethical considerations for study participants and restrictions of study resources.

Keywords

Clinical trials, personalized medicine, molecular diagnostic technique/biomarker, spatial epidemiology, genome data analysis, meta-analysis

Technologies

  • Data analysis
  • Statistical methodology for study scheme

Monographs, Papers and Articles

  • Matsui S, Buyse M, Simon R. Design and Analysis of Clinical Trials for Predictive Medicine. 2015, CRC Press.
  • Matsui S, Nonaka T, Choai Y. Design of phase III clinical trials with predictive biomarkers for personalized medicine. In Developments in Statistical Evaluation of Clinical Trials (eds K. van Montfort, J. Oud, W. Ghidey), 2014, Springer.
  • Matsui S, Choai Y, Nonaka T. Comparison of statistical analysis plans in randomize-all phase III trials with a predictive biomarker. Clinical Cancer Research 2014; 20;2820-30
  • Takahashi K, Nakao H, Hattori S. Cubic spline regression of J-shaped dose-response curves with likelihood-based assignments of grouped exposure levels. Journal of Biometrics & Biostatistics 2013; 4:181.
  • Matsui S, Simon R, Qu P, Shaughnessy JD, Barlogie B, Crowley J. Developing and validating continuous genomic signatures in randomized clinical trials for predictive medicine. Clinical Cancer Research 2012; 18: 6065-6073
  • Tango T, Takahashi K. A flexible spatial scan statistic with a restricted likelihood ratio for detecting disease clusters. Statistics in Medicine 2012; 31(30):4207-4218.