Skip to main content
. 2022 Feb 21;2021:546–555.

Figure 2:

Figure 2:

Visualization of the disease localization on the test images with ChexRadiNet and ground truth bounding boxes. The attention maps are generated from the final output tensor and overlapped on the original radiology images. The left image in each pair is the chest X-ray image and the right one is the generated attention map and the ground truth (in the yellow box).