Figure 6
From: High-resolution micro-CT for 3D infarct characterization and segmentation in mice stroke models
Deep learning methods for semi-automatic evaluation of the white fiber degradation in the TIA model and automatic infarct lesion segmentation, in the stroke model. In (A), we utilized a combination of manual segmentation of caudate putamen and automatic segmentation of white matter fibers via white top-hat transform. The thickness of the fibers was evaluated using the software VG Studio MAX 3.4 (VOLUME GRAPHICS GmbH, Heidelberg, Germany). In (B), an automatic deep learning-based segmentation workflow is proposed for the segmentation of the lesion penumbra and core. (C) The convolutional neural network (CNN) is utilized for automatic lesion detection. (D) Representative images of automatic lesion segmentation in the iodine-stained brain of mouse stroke model (tMCAO 45 min).
