LSTM-MA: A LSTM Method with Multi-Modality and Adjacency Constraint for Brain Image Segmentation
Kai Xie, Ying Wen · 2019
MR brain tissue segmentation is a significant problem in biomedical image processing. Inhomogeneous intensity and image noise influence the segmentation accuracy. In this paper, we propose a LSTM method with multi-modality and adjacency constraint for brain image segmentation, named LSTM-MA. Two feature sequence generation ways in our method are used, i.e., features with pixel-wise and superpixel-wise adjacency constraint. The LSTM model classifies the generated features into semantic labels to form the segmentation result. The evaluation experiments on BrainWeb and MRBrainS demonstrate that the proposed LSTM-MA with pixel-wise adjacency constraint achieves promising segmentation results, while LSTM-MA with superpixel-wise adjacency constraint shows its computational efficiency as well as robustness to noise.