Prognostic model of cancer patients using the relative survival regression tree

2014 Fiscal Year Final Research Report

• Project/Area Number: 24701042
• Research Category: Grant-in-Aid for Young Scientists (B)
• Allocation Type: Multi-year Fund
• Research Field: Cancer epidemiology and prevention
• Research Institution: Research Institute, Osaka Medical Center for Cancer and Cardiovascular Disaeses
• Principal Investigator: YURI Ito 地方独立行政法人大阪府立病院機構大阪府立成人病センター(研究所), その他部局等, その他 (60585305)
• Project Period (FY): 2012-04-01 – 2015-03-31
• Keywords: 生存解析 / がん登録 / 予後予測モデル

Outline of Final Research Achievements

We developed the regression tree based on the relative survival approach. Using this data mining approach, we can explore the prognostic factors and classify patients to significant subgroups from database with many candidate variables. We applied to the data of non-small-cell lung cancer patients from the hospital cancer registry data of our institute linking with the lifestyle survey for patients. Stage at diagnosis was the most important prognostic factor. Smoking status was a prognostic factor among the localised patients. We proposed that this approach can classified patients to subgroups with different prognosis.

Free Research Field

がん疫学