| Febrian Rachmadi |
| Thu 21 Feb 2019, 12:45 - 14:00 |
| IF, 4.31/33 |
If you have a question about this talk, please contact: Jodie Cameron (jcamero9)
Title: Predicting the Progression of White Matter Hyperintensities in Brain MRI using Generative Adversarial Networks and Irregularity Map
Abstract:
White matter hyperintensities (WMH) are common features in T2-FLAIR brain MRI and have been associated with stroke and the progression of dementia. One important aspect of WMH is its evolution over period of time. Previous studies have shown that the size (i.e., volume) of WMH on a patient may decrease (regress), stay the same, or increase (progress) over period of time. While factors that influence WMH evolution are still poorly understood, some factors (e.g., high WMH burden, hypertension, and increasing age) have been commonly associated. In this presentation, we will explore an end-to-end training model for predicting the evolution of WMH from baseline (year-0) to the following year (year-1) using generative adversarial networks (GANs) and irregularity map (IM).