Counterfactual Image Synthesis for Discovery of Personalized Predictive Image Markers

TL;DR

This paper introduces a counterfactual generative model to identify personalized imaging biomarkers predictive of disease progression, demonstrated on MS MRI data, revealing both known and novel markers.

Contribution

It presents a novel deep learning approach for counterfactual image synthesis to discover personalized predictive biomarkers in medical imaging.

Findings

Model produces counterfactuals reflecting established clinical markers.

Potential to discover novel, subject-specific predictive markers.

Effective on large-scale multi-center MS MRI dataset.

Abstract

The discovery of patient-specific imaging markers that are predictive of future disease outcomes can help us better understand individual-level heterogeneity of disease evolution. In fact, deep learning models that can provide data-driven personalized markers are much more likely to be adopted in medical practice. In this work, we demonstrate that data-driven biomarker discovery can be achieved through a counterfactual synthesis process. We show how a deep conditional generative model can be used to perturb local imaging features in baseline images that are pertinent to subject-specific future disease evolution and result in a counterfactual image that is expected to have a different future outcome. Candidate biomarkers, therefore, result from examining the set of features that are perturbed in this process. Through several experiments on a large-scale, multi-scanner, multi-center multiple sclerosis (MS) clinical trial magnetic resonance imaging (MRI) dataset of relapsing-remitting (RRMS) patients, we demonstrate that our model produces counterfactuals with changes in imaging features that reflect established clinical markers predictive of future MRI lesional activity at the population level. Additional qualitative results illustrate that our model has the potential to discover novel and subject-specific predictive markers of future activity.