Figure 7

Grad-CAM visualizations for an intermediate convolutional layer of the ResNet-50 incorporated in our SNN. Each column represents one image pair. The first four columns (a–d) show true positive classifications. The last two columns (e) and (f) depict a false positive and a false negative classification, respectively. The obtained attention maps clearly illustrate that the selected network layer focuses on the ribs and the outline of the thorax.