| abstract |
The invention discloses a cardiovascular calcification plaque detection system based on residual double attention mechanism. The present invention proposes a segmentation model based on the residual dual attention mechanism, including a backbone network module BackBone and a residual dual attention module, wherein the residual dual attention module includes a parallel first attention module and a second attention module, and a fusion module; the input of the two parallel attention modules is the output feature X of the BackBone module, and the output is the segmentation result. The invention solves the problem that the gradient disappears with the deepening of the network by constructing the residual double attention mechanism module, so that the problem that the features are completely retained with the layer-by-layer forward propagation is solved. Therefore, the new features can be more accurate and complete, and the segmented calcified regions are more accurate. |