Cluster Activation Mapping with Applications to Medical Imaging

Sarah M. Ryan, Iowa State University
Nichole Carlson, University of Colorado Anschutz Medical Campus
Harris Butler, University of Colorado Anschutz Medical Campus
Tasha Fingerlin, National Jewish Health
Lisa Maier, European Molecular Biology Laboratory
Fuyong Xing, University of Colorado Denver

This is a pre-print of the article Ryan, Sarah, Nichole Carlson, Harris Butler, Tasha Fingerlin, Lisa Maier, and Fuyong Xing. "Cluster Activation Mapping with Applications to Medical Imaging." arXiv preprint arXiv:2010.04794 (2020). Posted with permission.


An open question in deep clustering is how to understand what in the image is creating the cluster assignments. This visual understanding is essential to be able to trust the results of an inherently complex algorithm like deep learning, especially when the derived cluster assignments may be used to inform decision-making or create new disease sub-types. In this work, we developed novel methodology to generate CLuster Activation Mapping (CLAM) which combines an unsupervised deep clustering framework with a modification of Score-CAM, an approach for discriminative localization in the supervised setting. We evaluated our approach using a simulation study based on computed tomography scans of the lung, and applied it to 3D CT scans from a sarcoidosis population to identify new clusters of sarcoidosis based purely on CT scan presentation.