Learning to rank from medical imaging data

Fabian Pedregosa 1, 2 Elodie Cauvet 3 Gaël Varoquaux 1, 4 Christophe Pallier 1, 3 Bertrand Thirion 1 Alexandre Gramfort 1, 4
2 SIERRA - Statistical Machine Learning and Parsimony
DI-ENS - Département d'informatique de l'École normale supérieure, ENS Paris - École normale supérieure - Paris, INRIA Paris-Rocquencourt, CNRS : UMR8548
Abstract : Medical images can be used to predict a clinical score coding for the severity of a disease, a pain level or the complexity of a cognitive task. In all these cases, the predicted variable has a natural order. While a standard classifier discards this information, we would like to take it into account in order to improve prediction performance. A standard linear regression does model such information, however the linearity assumption is likely not be satisfied when predicting from pixel intensities in an image. In this paper we address these modeling challenges with a supervised learning procedure where the model aims to order or rank images. We use a linear model for its robustness in high dimension and its possible interpretation. We show on simulations and two fMRI datasets that this approach is able to predict the correct ordering on pairs of images, yielding higher prediction accuracy than standard regression and multiclass classification techniques.
Document type :
Conference papers
Third International Workshop on Machine Learning in Medical Imaging - MLMI 2012, Oct 2012, Nice, France. 2012

Contributor : Fabian Pedregosa <>
Submitted on : Sunday, September 30, 2012 - 2:11:59 PM
Last modification on : Friday, September 11, 2015 - 9:53:01 AM




  • HAL Id : hal-00717990, version 2
  • ARXIV : 1207.3598



Fabian Pedregosa, Elodie Cauvet, Gaël Varoquaux, Christophe Pallier, Bertrand Thirion, et al.. Learning to rank from medical imaging data. Third International Workshop on Machine Learning in Medical Imaging - MLMI 2012, Oct 2012, Nice, France. 2012. <hal-00717990v2>




Consultation de
la notice


Téléchargement du document