PCG-CAM: Enhanced class activation map using principal components of gradients and its applications in brain MRI

Lan Huang, Yangguang Shao, Wenju Hou, Hui Yang, Yan Wang, Nan Sheng, Yinglu Sun, Yao Wang

Abstract: Highlights•A method that locates tumor regions and provides visual explanations for diagnosis.•Leverage gradients to calculate feature maps’ weights for saliency maps.•Extract principal components from gradients to weight the feature maps.•Preserve more features to improve the completeness of saliency maps.•Provide more fine-grained pseudo-labels for weakly-supervised tasks.