Figure 2

Overview of segmentation pipeline. (a) Post-mortem high-field fluid-attenuated inversion recovery (FLAIR) MRI corresponding to the tissue slab used for light sheet fluorescence microscopy. (b) This image illustrates the manual segmentation result as ground truth. (c) This image shows the segmentation outcome known as prediction for grey matter (GM), normal-appearing white matter (NAWM) and white matter hyperintensity (WMH). (d) After successful segmentation of grey and white matter, we applied KMeans clustering algorithm to (sub-)segment both NAWM and WMH based on changes on FLAIR signal. Scale bars, 1 cm (a–d).