Full-automatic PopPK modeling utilizing an evolutionary algorithm

2014 Fiscal Year Final Research Report

• Project/Area Number: 25670073
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Multi-year Fund
• Research Field: Medical pharmacy
• Research Institution: Kyoto University
• Principal Investigator: YAMASHITA Fumiyoshi 京都大学, 薬学研究科(研究院), 准教授 (30243041)
• Co-Investigator(Renkei-kenkyūsha): YANO Ikuko 京都大学, 薬学研究科, 准教授 (50273446) ; AMANO Akira 立命館大学, 生命科学部, 教授 (60252491)
• Project Period (FY): 2013-04-01 – 2015-03-31
• Keywords: 母集団薬物動態解析 / 共変量モデル / 遺伝的アルゴリズム / 遺伝的数式プログラミング / 個体間変動 / 拡張最小二乗法

Outline of Final Research Achievements

A population pharmacokinetic model is composed of average pharmacokinetic parameters, their inter- and intra-individual variabilities. In order to predict blood concentration profile of a drug in each individual, we have to develop an accurate covariate model that represents quantitative relationship between pharmacokinetic parameters and patient information. However, it is a labor-intensive process because there are too many possibilities comprising covariate candidates. In the present study a full-automatic system for covariate modeling was developed, that utilizes an algorithm of evolving mathematical formulae on computers.

Free Research Field

薬物動態学