arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2007.00453cs.CVcs.LG
收藏

M3d-CAM:一个PyTorch库,用于生成用于医学深度学习的3D数据注意力图

M3d-CAM: A PyTorch library to generate 3D data attention maps for medical deep learning

Technical university of Darmstadt(达姆施塔特工业大学);Universitätsklinikum Frankfurt(法兰克福大学医院)

Karol Gotkowski, Camila Gonzalez, Andreas Bucher, Anirban Mukhopadhyay

摘要:

M3 d-CAM是一个易于使用的库,用于生成基于CNN的PyTorch模型的注意力地图,提高了人类模型预测的可解释性。注意力地图可以使用多种方法生成,如Guided Backpropagation,Grad-CAM,Guided Grad-CAM和Grad-CAM++。这些注意力图可视化了输入数据中在某个层对模型预测影响最大的区域。此外,M3 d-CAM支持2D和3D数据用于分类和分割任务。一个关键的特性是在大多数情况下,只需要一行代码就可以为模型生成注意力地图,使M3 d-CAM基本上即插即用。

英文摘要:

M3d-CAM is an easy to use library for generating attention maps of CNN-based PyTorch models improving the interpretability of model predictions for humans. The attention maps can be generated with multiple methods like Guided Backpropagation, Grad-CAM, Guided Grad-CAM and Grad-CAM++. These attention maps visualize the regions in the input data that influenced the model prediction the most at a certain layer. Furthermore, M3d-CAM supports 2D and 3D data for the task of classification as well as for segmentation. A key feature is also that in most cases only a single line of code is required for generating attention maps for a model making M3d-CAM basically plug and play.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑