Psychotropic drug classification based on sleep-wake behaviour of rats

Authors: Wouters, Kristien¹; Ahnaou, Abdellah²; Abrahantes, Jose Cortinas¹; Molenberghs, Geert¹; Geys, Helena³; Bijnens, Luc²; Drinkenburg, Wilhelmus H. I. M.²

Source: Journal of the Royal Statistical Society: Series C (Applied Statistics), Volume 56, Number 2, March 2007 , pp. 223-234(12)

Publisher: Wiley-Blackwell

Buy & download fulltext article:

Price: $48.00 plus tax (Refund Policy)

Abstract:

Summary.

The aim of the paper is to present methodology for the classification of potential psychotropic drugs on the basis of their activity. We first sketch the background of this class of drugs and then zoom in on so-called pharmacoelectroencephalogram studies. These data pose some statistical challenges. For classification purposes, we propose a flexible hierarchical discriminant analysis tool, allowing us to take the specific nature of the drug class into account, as well as the features of the mixed models, in combination with fractional polynomials, fitted to the electroencephalogram data. The method is evaluated against the background of existing methods. The method's performance is studied by using a comprehensive analysis of a large electroencephalogram data set.

Keywords: Discriminant analysis; Electroencephalogram; Fractional polynomials; Longitudinal data; Mixed model

Document Type: Research article

DOI: http://dx.doi.org/10.1111/j.1467-9876.2007.00575.x

Affiliations: 1: Universiteit Hasselt, Diepenbeek, Belgium 2: Johnson & Johnson Pharmaceutical Research and Development, Beerse, Belgium 3: Johnson & Johnson Pharmaceutical Research and Development, Beerse, and Universiteit Hasselt, Diepenbeek, Belgium

Publication date: 2007-03-01