Development of multivariable graph representation data analysis method and its application to clinical data

2020 Fiscal Year Final Research Report

• Project/Area Number: 17K18445
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Statistical science; Epidemiology and preventive medicine
• Research Institution: Shizuoka Prefectural Hospital Organization (2019-2020); Foundation for Biomedical Research and Innovation at Kobe (2018); Osaka University (2017)
• Principal Investigator: Nakatani Eiji 地方独立行政法人静岡県立病院機構静岡県立総合病院(救急診療部、循環器病診療部、がん診療部、臨床診療部, 統計解析室, 室長 (80627670)
• Project Period (FY): 2017-04-01 – 2021-03-31
• Keywords: 臨床経過 / グラフ表現図 / 構造化モデル / アンサンブル学習 / 集合解 / 樹木モデル / 平均因果効果 / 交互作用項検定

Outline of Final Research Achievements

After the onset of a disease, the disease progresses with a variety of symptoms. I conducted a study to draw a statistically accurate picture of the typical symptom changes in patients with such a disease using data. In the first half of the study, I listed the statistical problems for drawing such a diagram. In the second half of the study, I conducted individual statistical studies on these problems. Through this research, I published many papers in the field of medicine and gave four international conference presentations in the field of statistics.

Free Research Field

医療統計学，疫学

Academic Significance and Societal Importance of the Research Achievements

この研究において進めた，疾患に罹った患者における代表的な症状の移り変わりを示す図を統計学に正確に描ければ，医療における治療やケア計画の立案に役立つし，疾患の特徴をより深く理解することができます．まだ研究は途中ではありますが，この研究のプロセスでわかった統計学や医療分野での問題点に対する研究の成果は，多くの論文公表と学会発表を行ったことで，医療統計学や疫学の分野での発展に寄与したと考えています．