Khakzar A, Zhang Y, Mansour W, Cai Y, Li Y, Zhang Y, Kim ST, Navab N (2021)
Publication Type: Conference contribution
Publication year: 2021
Publisher: Springer Science and Business Media Deutschland GmbH
Book Volume: 12903 LNCS
Pages Range: 391-401
Conference Proceedings Title: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Event location: Virtual, Online
ISBN: 9783030871987
DOI: 10.1007/978-3-030-87199-4_37
Neural networks have demonstrated remarkable performance in classification and regression tasks on chest X-rays. In order to establish trust in the clinical routine, the networks’ prediction mechanism needs to be interpretable. One principal approach to interpretation is feature attribution. Feature attribution methods identify the importance of input features for the output prediction. Building on Information Bottleneck Attribution (IBA) method, for each prediction we identify the chest X-ray regions that have high mutual information with the network’s output. Original IBA identifies input regions that have sufficient predictive information. We propose Inverse IBA to identify all informative regions. Thus all predictive cues for pathologies are highlighted on the X-rays, a desirable property for chest X-ray diagnosis. Moreover, we propose Regression IBA for explaining regression models. Using Regression IBA we observe that a model trained on cumulative severity score labels implicitly learns the severity of different X-ray regions. Finally, we propose Multi-layer IBA to generate higher resolution and more detailed attribution/saliency maps. We evaluate our methods using both human-centric (ground-truth-based) interpretability metrics, and human-agnostic feature importance metrics on NIH Chest X-ray8 and BrixIA datasets. The code (https://github.com/CAMP-eXplain-AI/CheXplain-IBA ) is publicly available.
APA:
Khakzar, A., Zhang, Y., Mansour, W., Cai, Y., Li, Y., Zhang, Y.,... Navab, N. (2021). Explaining COVID-19 and Thoracic Pathology Model Predictions by Identifying Informative Input Features. In Marleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, Caroline Essert (Eds.), Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (pp. 391-401). Virtual, Online: Springer Science and Business Media Deutschland GmbH.
MLA:
Khakzar, Ashkan, et al. "Explaining COVID-19 and Thoracic Pathology Model Predictions by Identifying Informative Input Features." Proceedings of the 24th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2021, Virtual, Online Ed. Marleen de Bruijne, Philippe C. Cattin, Stéphane Cotin, Nicolas Padoy, Stefanie Speidel, Yefeng Zheng, Caroline Essert, Springer Science and Business Media Deutschland GmbH, 2021. 391-401.
BibTeX: Download