4D Segmentation of Longitudinal Brain MR Images with Consistent Cortical Thickness Measurement

Accurate segmentation of the brain MR images plays an important role in investigation of neurodegenerative changes in the cerebral cortex. However, most of the previous algorithms were proposed for segmentation of 3D images and few studies have taken the temporal consistency of cortical-thickness changes into account during the longitudinal studies. In this paper, we propose a 4D segmentation framework for the adult brain MR images with consistent longitudinal cortical thickness changes. Specifically, we utilize local intensity information to address the intensity inhomogeneity, spatial cortical thickness constraint to maintain the cortical thickness within a reasonable range, and temporal cortical thickness constraint to ensure the cortical thickness at the current timepoint to be temporally consistent with thicknesses in the neighboring time-points. The proposed method has been tested on BLSA dataset and ADNI dataset. Both qualitative and quantitative experimental results demonstrate the accuracy and consistency of the proposed method, in comparison to other state-of-the-art 4D segmentation methods.