APAUNet: Axis Projection Attention UNet for Small Target in 3D Medical Segmentation
This addresses the challenge of accurately segmenting small lesions in 3D medical scans for improved diagnosis, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles the problem of small target segmentation in 3D medical images by proposing APAUNet, which uses axis projection attention to capture contextual information from different views, achieving average dice scores of 87.84 on BTCV, 84.48 on MSD-Liver, and 69.13 on MSD-Pancreas, outperforming previous methods.
In 3D medical image segmentation, small targets segmentation is crucial for diagnosis but still faces challenges. In this paper, we propose the Axis Projection Attention UNet, named APAUNet, for 3D medical image segmentation, especially for small targets. Considering the large proportion of the background in the 3D feature space, we introduce a projection strategy to project the 3D features into three orthogonal 2D planes to capture the contextual attention from different views. In this way, we can filter out the redundant feature information and mitigate the loss of critical information for small lesions in 3D scans. Then we utilize a dimension hybridization strategy to fuse the 3D features with attention from different axes and merge them by a weighted summation to adaptively learn the importance of different perspectives. Finally, in the APA Decoder, we concatenate both high and low resolution features in the 2D projection process, thereby obtaining more precise multi-scale information, which is vital for small lesion segmentation. Quantitative and qualitative experimental results on two public datasets (BTCV and MSD) demonstrate that our proposed APAUNet outperforms the other methods. Concretely, our APAUNet achieves an average dice score of 87.84 on BTCV, 84.48 on MSD-Liver and 69.13 on MSD-Pancreas, and significantly surpass the previous SOTA methods on small targets.