The development of statistical models based on function data for the high-dimensional medical data

2017 Fiscal Year Final Research Report

• Project/Area Number: 26730023
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science
• Research Institution: Shizuoka University (2015-2017); Kurume University (2014)
• Principal Investigator: Araki Yuko 静岡大学, 情報学部, 准教授 (80403913)
• Project Period (FY): 2014-04-01 – 2018-03-31
• Keywords: 関数データ解析 / 高次元データ / 判別モデル / 情報量規準 / 非線形モデル / 多変量データ解析 / スパース推定 / 構造方程式モデル

Outline of Final Research Achievements

In this research, we have developed statistical models to extract useful information from super-high dimensional data without loss of information by creating some new dimension reduction techniques. We constructed a classification model which detects onset of some disease on early stage based on MRI data. Further, we elucidated a mechanism between brain structure, levels of activity and clinical endpoints. In addition, the survival model with sparse constraints were applied to the long term and large sample size follow up data aimed at realizing health longevity society in Japan. All the results described above were presented at both domestic and international conferences. We had further worked out for constructing functional structural equation modeling(SEM). However, since there were some issues which need further consideration regarding to characteristics of estimator, the SEM modeling was still developing.

Free Research Field

統計科学