01 Oct 21. Featured Paper

Probabilistic deep learning with adversarial training & volume interval estimation - better ways to perform & evaluate predictive models for white matter hyperintensities evolution.

Link to paper on Springer Nature

 

Authors

Muhammad Febrian Rachmadi, Maria del C. Valdés-Hernández, Rizal Maulana, Joanna Wardlaw, Stephen Makin, Henrik Skibbe

 

Abstract

Predicting disease progression always involves a high degree of uncertainty.

White matter hyperintensities (WMHs) are the main neuroradiological feature of small vessel disease & a common finding in brain scans of dementia patients & older adults.

In predicting their progression previous studies have identified two main challenges: 1) uncertainty in predicting the areas/boundaries of shrinking & growing WMHs & 2) uncertainty in the estimation of future WMHs volume.

This study proposes the use of a probabilistic deep learning model called Probabilistic U-Net trained with adversarial loss for capturing & modelling spatial uncertainty in brain MR images.

This study also proposes an evaluation procedure named volume interval estimation (VIE) for improving the interpretation of & confidence in the predictive deep learning model.

Our experiments show that the Probabilistic U-Net with adversarial training improved the performance of non-probabilistic U-Net in Dice similarity coefficient for predicting the areas of shrinking WMHs, growing WMHs, stable WMHs, & their average by up to 3.35%, 2.94%, 0.47%, & 1.03% respectively.

It also improved the volume estimation by 11.84% in the “Correct Prediction in Estimated Volume Interval” metric as per the newly proposed VIE evaluation procedure.

 

Keywords
  • Progression prediction
  • Volume interval estimation
  • White matter hyperintensities

 

 

Social media tags & titles

Featured paper: Probabilistic deep learning with adversarial training & volume interval estimation - better ways to perform & evaluate predictive models for white matter hyperintensities evolution

@SVDResearch