Classification of high dimensional data for cervical cancer detection

Charles Bouveyron (1) , Camille Brunet (2) , Vincent Vigneron (2)

1 CES - Centre d'économie de la Sorbonne
2 IBISC - Informatique, Biologie Intégrative et Systèmes Complexes

In this paper, the performance of different generative methods for the classification of cervical nuclei are compared in order to detect cancer of cervix. These methods include classical Bayesian approaches, such as Linear Discriminant Analysis (LDA), Quadratic Discriminant Analysis (QDA) or Mixture Discriminant Analysis (MDA) and a high-dimensional approach (HDDA) recently developed. The classification of cervical nuclei presents 2 main statistical issues, scarce population and high-dimensional data, which impact on the ability to successfully discriminate the different classes. This paper presents an approach to face the problems of unbalanced data and high-dimensions in the context of cervical cancer detection.

Domaines: Applications [stat.AP]

Charles Bouveyron, Camille Brunet, Vincent Vigneron. Classification of high dimensional data for cervical cancer detection. 17th European Symposium on Artificial Neural Networks (ESANN 2009), Apr 2009, Bruges, Belgium. ⟨hal-00394431⟩