An Adaptive 3D U-Net for White Matter, Gray Matter and Cerebrospinal Fluid Segmentation from 3D-Brain MRI
Pham The Bao, ANH TUAN TRAN, ANH TUAN(A) TRAN, NHI LAM THUY LE, Jin Young Kim · 2021
Many methods for Alzheimer's disease detection in brain magnetic resonance imaging (MRI) is related to white matter (WM), grey matter (GM), and cerebrospinal fluid (CSF) regions. Therefore, in many neurological applications, the segmentation of these tissues in magnetic resonance imaging (MRI) plays an important role in the analysis. In the trend of deep learning, an application using 3 dimensional (3D) Convolution Neural Network will help doctors get the best result segmentation. In this research, we proposed an effective approach to segment automatically these tissues in Brain MRI 3D by using an adaptive 3D U-Net. In the experiments, a real MRI database, The Internet Brain Segmentation Repository (IBSR) 18, is evaluated with the proposed method and gives the promising Dice with 0.92, 0.87, 0.81 for WM, GM, and CSF segmentation.