Thorax Disease Classification with Attention Guided Convolutional Neural Network
Pattern Recognition Letters(2020)
摘要
•A novel attention guided convolutional neural network (AG-CNN) is proposed for thorax disease identification.•AG-CNN integrates the global and local information to improve the recognition performance.•We introduce a CNN training baseline, which produces competitive results to the state-of-the-art methods.•A new state-of-the-art is yielded by AG-CNN on the ChestX-ray14 dataset.
更多查看译文
关键词
CXR image classification,Visual attention,Feature ensemble
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络