Hippocampus Segmentation Method Applying Coordinate Attention Mechanism and Dynamic Convolution Network

Precisely segmenting the hippocampus from brain is crucial for diagnosing neurodegenerative illnesses such as Alzheimer’s disease, depression, etc. In this research, we propose an enhanced segmentation algorithm based on 3D U-Net that can significantly increase performance. First, a dynamic convolution block designed to extract information more comprehensively in steps of U-Net’s encoder and decoder. addition, improved coordinate attention applied skip connections step weight reduce redundancy other unimportant location information. The proposed work uses soft pooling methods instead max loss during downsampling steps. datasets employed research were obtained MICCAI 2013 SATA Challenge (MICCAI) Harmonized Protocol initiative Disease Neuroimaging Initiative (HarP). experimental results two prove outperforms commonly used algorithms. On HarP, dice by 3.52%, mIoU increases 2.65%, F1 score 3.38% contrast baseline. MICCAI, dice, mIoU, 1.13%, 0.85%, 1.08%, respectively. Overall, model common