Quantifying the Value of Lateral Views in Deep Learning for Chest X-rays

Mohammad Hashir, Hadrien Bertrand, Joseph Paul Cohen
Proceedings of the Third Conference on Medical Imaging with Deep Learning, PMLR 121:288-303, 2020.

Abstract

Most deep learning models in chest X-ray prediction utilize the posteroanterior (PA) view due to the lack of other views available. PadChest is a large-scale chest X-ray dataset that has almost 200 labels and multiple views available. In this work, we use PadChest to explore multiple approaches to merging the PA and lateral views for predicting the radiological labels associated with the X-ray image. We find that different methods of merging the model utilize the lateral view differently. We also find that including the lateral view increases performance for 32 labels in the dataset, while being neutral for the others. The increase in overall performance is comparable to the one obtained by using only the PA view with twice the amount of patients in the training set.

Cite this Paper


BibTeX
@InProceedings{pmlr-v121-hashir20a, title = {Quantifying the Value of Lateral Views in Deep Learning for Chest X-rays}, author = {Hashir, Mohammad and Bertrand, Hadrien and Cohen, Joseph Paul}, booktitle = {Proceedings of the Third Conference on Medical Imaging with Deep Learning}, pages = {288--303}, year = {2020}, editor = {Arbel, Tal and Ben Ayed, Ismail and de Bruijne, Marleen and Descoteaux, Maxime and Lombaert, Herve and Pal, Christopher}, volume = {121}, series = {Proceedings of Machine Learning Research}, month = {06--08 Jul}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v121/hashir20a/hashir20a.pdf}, url = { http://proceedings.mlr.press/v121/hashir20a.html }, abstract = {Most deep learning models in chest X-ray prediction utilize the posteroanterior (PA) view due to the lack of other views available. PadChest is a large-scale chest X-ray dataset that has almost 200 labels and multiple views available. In this work, we use PadChest to explore multiple approaches to merging the PA and lateral views for predicting the radiological labels associated with the X-ray image. We find that different methods of merging the model utilize the lateral view differently. We also find that including the lateral view increases performance for 32 labels in the dataset, while being neutral for the others. The increase in overall performance is comparable to the one obtained by using only the PA view with twice the amount of patients in the training set.} }
Endnote
%0 Conference Paper %T Quantifying the Value of Lateral Views in Deep Learning for Chest X-rays %A Mohammad Hashir %A Hadrien Bertrand %A Joseph Paul Cohen %B Proceedings of the Third Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2020 %E Tal Arbel %E Ismail Ben Ayed %E Marleen de Bruijne %E Maxime Descoteaux %E Herve Lombaert %E Christopher Pal %F pmlr-v121-hashir20a %I PMLR %P 288--303 %U http://proceedings.mlr.press/v121/hashir20a.html %V 121 %X Most deep learning models in chest X-ray prediction utilize the posteroanterior (PA) view due to the lack of other views available. PadChest is a large-scale chest X-ray dataset that has almost 200 labels and multiple views available. In this work, we use PadChest to explore multiple approaches to merging the PA and lateral views for predicting the radiological labels associated with the X-ray image. We find that different methods of merging the model utilize the lateral view differently. We also find that including the lateral view increases performance for 32 labels in the dataset, while being neutral for the others. The increase in overall performance is comparable to the one obtained by using only the PA view with twice the amount of patients in the training set.
APA
Hashir, M., Bertrand, H. & Cohen, J.P.. (2020). Quantifying the Value of Lateral Views in Deep Learning for Chest X-rays. Proceedings of the Third Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 121:288-303 Available from http://proceedings.mlr.press/v121/hashir20a.html .

Related Material