Deep Hierarchical Multi-label Classification of Chest X-ray Images

Haomin Chen, Shun Miao, Daguang Xu, Gregory D. Hager, Adam P. Harrison
Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning, PMLR 102:109-120, 2019.

Abstract

Chest X-rays (CXRs) are a crucial and extraordinarily common diagnostic tool, leading to heavy research for Computer-Aided Diagnosis (CAD) solutions. However, both high classification accuracy and meaningful model predictions that respect and incorporate clinical taxonomies are crucial for CAD usability. To this end, we present a deep Hierarchical Multi-Label Classification (HMLC) approach for CXR CAD. Different than other hierarchical systems, we show that first training the network to model conditional probability directly and then refining it with unconditional probabilities is key in boosting performance. In addition, we also formulate a numerically stable cross-entropy loss function for unconditional probabilities that provides concrete performance improvements. To the best of our knowledge, we are the first to apply HMLC to medical imaging CAD. We extensively evaluate our approach on detecting 14 abnormality labels from the PLCO dataset, which comprises 198,000 manually annotated CXRs. We report a mean Area Under the Curve (AUC) of 0.887, the highest yet reported for this dataset. These performance improvements, combined with the inherent usefulness of taxonomic predictions, indicate that our approach represents a useful step forward for CXR CAD.

Cite this Paper

BibTeX
@InProceedings{pmlr-v102-chen19a, title = {Deep Hierarchical Multi-label Classification of Chest X-ray Images}, author = {Chen, Haomin and Miao, Shun and Xu, Daguang and Hager, Gregory D. and Harrison, Adam P.}, booktitle = {Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning}, pages = {109--120}, year = {2019}, editor = {Cardoso, M. Jorge and Feragen, Aasa and Glocker, Ben and Konukoglu, Ender and Oguz, Ipek and Unal, Gozde and Vercauteren, Tom}, volume = {102}, series = {Proceedings of Machine Learning Research}, month = {08--10 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v102/chen19a/chen19a.pdf}, url = {https://proceedings.mlr.press/v102/chen19a.html}, abstract = {Chest X-rays (CXRs) are a crucial and extraordinarily common diagnostic tool, leading to heavy research for Computer-Aided Diagnosis (CAD) solutions. However, both high classification accuracy and meaningful model predictions that respect and incorporate clinical taxonomies are crucial for CAD usability. To this end, we present a deep Hierarchical Multi-Label Classification (HMLC) approach for CXR CAD. Different than other hierarchical systems, we show that first training the network to model conditional probability directly and then refining it with unconditional probabilities is key in boosting performance. In addition, we also formulate a numerically stable cross-entropy loss function for unconditional probabilities that provides concrete performance improvements. To the best of our knowledge, we are the first to apply HMLC to medical imaging CAD. We extensively evaluate our approach on detecting 14 abnormality labels from the PLCO dataset, which comprises 198,000 manually annotated CXRs. We report a mean Area Under the Curve (AUC) of 0.887, the highest yet reported for this dataset. These performance improvements, combined with the inherent usefulness of taxonomic predictions, indicate that our approach represents a useful step forward for CXR CAD.} }
Endnote
%0 Conference Paper %T Deep Hierarchical Multi-label Classification of Chest X-ray Images %A Haomin Chen %A Shun Miao %A Daguang Xu %A Gregory D. Hager %A Adam P. Harrison %B Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2019 %E M. Jorge Cardoso %E Aasa Feragen %E Ben Glocker %E Ender Konukoglu %E Ipek Oguz %E Gozde Unal %E Tom Vercauteren %F pmlr-v102-chen19a %I PMLR %P 109--120 %U https://proceedings.mlr.press/v102/chen19a.html %V 102 %X Chest X-rays (CXRs) are a crucial and extraordinarily common diagnostic tool, leading to heavy research for Computer-Aided Diagnosis (CAD) solutions. However, both high classification accuracy and meaningful model predictions that respect and incorporate clinical taxonomies are crucial for CAD usability. To this end, we present a deep Hierarchical Multi-Label Classification (HMLC) approach for CXR CAD. Different than other hierarchical systems, we show that first training the network to model conditional probability directly and then refining it with unconditional probabilities is key in boosting performance. In addition, we also formulate a numerically stable cross-entropy loss function for unconditional probabilities that provides concrete performance improvements. To the best of our knowledge, we are the first to apply HMLC to medical imaging CAD. We extensively evaluate our approach on detecting 14 abnormality labels from the PLCO dataset, which comprises 198,000 manually annotated CXRs. We report a mean Area Under the Curve (AUC) of 0.887, the highest yet reported for this dataset. These performance improvements, combined with the inherent usefulness of taxonomic predictions, indicate that our approach represents a useful step forward for CXR CAD.
APA
Chen, H., Miao, S., Xu, D., Hager, G.D. & Harrison, A.P.. (2019). Deep Hierarchical Multi-label Classification of Chest X-ray Images. Proceedings of The 2nd International Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 102:109-120 Available from https://proceedings.mlr.press/v102/chen19a.html.

Related Material