RT Journal Article
SR Electronic
T1 Prediction of white matter hyperintensities evolution one-year post-stroke from a single-point brain MRI and stroke lesions information
JF bioRxiv
FD Cold Spring Harbor Laboratory
SP 2022.12.14.520239
DO 10.1101/2022.12.14.520239
A1 Rachmadi, Muhammad Febrian
A1 Valdés-Hernández, Maria del C.
A1 Makin, Stephen
A1 Wardlaw, Joanna
A1 Skibbe, Henrik
YR 2023
UL http://biorxiv.org/content/early/2023/10/16/2022.12.14.520239.abstract
AB Predicting the evolution of white matter hyperintensities (WMH), a common feature in brain magnetic resonance imaging (MRI) scans of older adults (i.e., whether WMH will grow, remain stable, or shrink with time) is important for personalised therapeutic interventions. However, this task is difficult mainly due to the myriad of vascular risk factors and comorbidities that influence it, and the low specificity and sensitivity of the image intensities and textures alone for predicting WMH evolution. Given the predominantly vascular nature of WMH, in this study, we evaluate the impact of incorporating stroke lesion information to a probabilistic deep learning model to predict the evolution of WMH 1-year after the baseline image acquisition, taken soon after a mild stroke event, using T2-FLAIR brain MRI. The Probabilistic U-Net was chosen for this study due to its capability of simulating and quantifying the uncertainties involved in the prediction of WMH evolution. We propose to use an additional loss called volume loss to train our model, and incorporate stroke lesions information, an influential factor in WMH evolution. Our experiments showed that jointly segmenting the disease evolution map (DEM) of WMH and stroke lesions, improved the accuracy of the DEM representing WMH evolution. The combination of introducing the volume loss and joint segmentation of DEM of WMH and stroke lesions outperformed other model configurations with mean volumetric absolute error of 0.0092 ml (down from 1.7739 ml) and 0.47% improvement on average Dice similarity coefficient in shrinking, growing and stable WMH.Competing Interest StatementThe authors have declared no competing interest.