Learning to rank from medical imaging data

TL;DR

This paper introduces a supervised learning approach for ranking medical images based on their clinical relevance, leveraging linear models to improve prediction accuracy over traditional methods.

Contribution

It proposes a novel ranking-based modeling framework for medical imaging data that captures the ordinal nature of clinical scores, outperforming standard regression and classification.

Findings

Higher accuracy in predicting image orderings compared to standard methods

Effective in high-dimensional settings with pixel intensity data

Validated on simulated and real fMRI datasets

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.