To addresses the issues of low contrast and inaccurate segmentation of tiny vessels in retinal images,a U-shaped network incorporating multi-level residuals and three-space attention mechanism is proposed.In encoding stage,a multi-level residual module is added after inputting original images for preserving image features,and additionally,batch normalization and Dropout are integrated into the residual module to prevent vanishing gradient and feature data redundancy within the deep network.In decoding stage,a three-space attention mechanism is adopted to assign different weights to the features from the original images,down-sampled images,and up-sampled images,thus enhancing feature texture and position information,and achieving precise segmentation of tiny blood vessels.Experimental results on a public color fundus image dataset demonstrate that the proposed algorithm achieves higher accuracy(0.985),specificity(0.991),sensitivity(0.829),and AUC(0.985)than the existing algorithms.Moreover,the vessel maps obtained by the comparison with the gold standard are of significant reference value in clinic.

To address the challenge of unclear features and difficulties in pulmonary nodule CT image detection,an efficient attention feature pyramid network is proposed.The network firstly employs a feature pyramid of multi-scale feature fusion as the backbone network for effectively preserving both low-and high-level features,and uses the depthwise separable convolutional neural network to extract feature information.Then,the attention mechanism is integrated into the backbone network for assigning weights to salient feature information.Finally,the proposed algorithm is applied to Lung-PET-CT-Dx dataset and Luna16 dataset,and the experimental results demonstrate that the proposed algorithm has higher precision,recall rate and mAP value than the existing comparative algorithms,substantiating its superiority in pulmonary nodule detection.