Self-Supervised Pretraining with DICOM metadata in Ultrasound Imaging

Szu-Yen Hu, Shuhang Wang, Wei-Hung Weng, JingChao Wang, XiaoHong Wang, Arinc Ozturk, Quan Li, Viksit Kumar, Anthony E. Samir
Proceedings of the 5th Machine Learning for Healthcare Conference, PMLR 126:732-749, 2020.

Abstract

Modern deep learning algorithms geared towards clinical adaption usually rely on a large amount of high fidelity labeled data. Low-resource settings pose challenges like acquiring high fidelity data and becomes the bottleneck for developing artificial intelligence applications. Ultrasound images, stored in Digital Imaging and Communication in Medicine (DICOM) format, have additional metadata data corresponding to ultrasound image parameters and medical exams. In this work, we leverage DICOM metadata from ultrasound images to help learn representations of the ultrasound image. We demonstrate that the proposed method outperforms the approaches without using metadata across a variety of downstream tasks.

Cite this Paper

BibTeX
@InProceedings{pmlr-v126-hu20a, title = {Self-Supervised Pretraining with DICOM metadata in Ultrasound Imaging}, author = {Hu, Szu-Yen and Wang, Shuhang and Weng, Wei-Hung and Wang, JingChao and Wang, XiaoHong and Ozturk, Arinc and Li, Quan and Kumar, Viksit and Samir, Anthony E.}, booktitle = {Proceedings of the 5th Machine Learning for Healthcare Conference}, pages = {732--749}, year = {2020}, editor = {Doshi-Velez, Finale and Fackler, Jim and Jung, Ken and Kale, David and Ranganath, Rajesh and Wallace, Byron and Wiens, Jenna}, volume = {126}, series = {Proceedings of Machine Learning Research}, month = {07--08 Aug}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v126/hu20a/hu20a.pdf}, url = {https://proceedings.mlr.press/v126/hu20a.html}, abstract = {Modern deep learning algorithms geared towards clinical adaption usually rely on a large amount of high fidelity labeled data. Low-resource settings pose challenges like acquiring high fidelity data and becomes the bottleneck for developing artificial intelligence applications. Ultrasound images, stored in Digital Imaging and Communication in Medicine (DICOM) format, have additional metadata data corresponding to ultrasound image parameters and medical exams. In this work, we leverage DICOM metadata from ultrasound images to help learn representations of the ultrasound image. We demonstrate that the proposed method outperforms the approaches without using metadata across a variety of downstream tasks.} }
Endnote
%0 Conference Paper %T Self-Supervised Pretraining with DICOM metadata in Ultrasound Imaging %A Szu-Yen Hu %A Shuhang Wang %A Wei-Hung Weng %A JingChao Wang %A XiaoHong Wang %A Arinc Ozturk %A Quan Li %A Viksit Kumar %A Anthony E. Samir %B Proceedings of the 5th Machine Learning for Healthcare Conference %C Proceedings of Machine Learning Research %D 2020 %E Finale Doshi-Velez %E Jim Fackler %E Ken Jung %E David Kale %E Rajesh Ranganath %E Byron Wallace %E Jenna Wiens %F pmlr-v126-hu20a %I PMLR %P 732--749 %U https://proceedings.mlr.press/v126/hu20a.html %V 126 %X Modern deep learning algorithms geared towards clinical adaption usually rely on a large amount of high fidelity labeled data. Low-resource settings pose challenges like acquiring high fidelity data and becomes the bottleneck for developing artificial intelligence applications. Ultrasound images, stored in Digital Imaging and Communication in Medicine (DICOM) format, have additional metadata data corresponding to ultrasound image parameters and medical exams. In this work, we leverage DICOM metadata from ultrasound images to help learn representations of the ultrasound image. We demonstrate that the proposed method outperforms the approaches without using metadata across a variety of downstream tasks.
APA
Hu, S., Wang, S., Weng, W., Wang, J., Wang, X., Ozturk, A., Li, Q., Kumar, V. & Samir, A.E.. (2020). Self-Supervised Pretraining with DICOM metadata in Ultrasound Imaging. Proceedings of the 5th Machine Learning for Healthcare Conference, in Proceedings of Machine Learning Research 126:732-749 Available from https://proceedings.mlr.press/v126/hu20a.html.

Related Material