Effect of image resolution on automated classification of chest X-rays
Journal of Medical Imaging · 04.08.2023 · 10 (04) · 1-16
DOI: 10.1117/1.jmi.10.4.044503 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Medical Imaging, erschienen am 04.08.2023. Verfasst von Md Inzamam Ul Haque und 5 weiteren Autor:innen. Beteiligte Einrichtung: University of Tennessee, The Bredesen Center, Knoxville, Tennessee, United States. Die Arbeit wurde bisher 28-mal zitiert (Crossref) und nennt 39 Literaturangaben. Der Volltext ist frei in PubMed Central verfügbar. Gefördert von: NCATS NIH HHS.
Abstract (Originalsprache)
PurposeDeep learning (DL) models have received much attention lately for their ability to achieve expert-level performance on the accurate automated analysis of chest X-rays (CXRs). Recently available public CXR datasets include high resolution images, but state-of-the-art models are trained on reduced size images due to limitations on graphics processing unit memory and training time. As computing hardware continues to advance, it has become feasible to train deep convolutional neural networks on high-resolution images without sacrificing detail by downscaling. This study examines the effect of increased resolution on CXR classification performance.ApproachWe used the publicly available MIMIC-CXR-JPG dataset, comprising 377,110 high resolution CXR images for this study. We applied image downscaling from native resolution to 2048 × 2048 pixels, 1024 × 1024 pixels, 512 × 512 pixels, and 256 × 256 pixels and then we used the DenseNet121 and EfficientNet-B4 DL models to evaluate clinical task performance using these four downscaled image resolutions.ResultsWe find that while some clinical findings are more reliably labeled using high resolutions, many other findings are actually labeled better using downscaled inputs. We qualitatively verify that tasks requiring a large receptive field are better suited to downscaled low resolution input images, by inspecting effective receptive fields and class activation maps of trained models. Finally, we show that stacking an ensemble across resolutions outperforms each individual learner at all input resolutions while providing interpretable scale weights, indicating that diverse information is extracted across resolutions.ConclusionsThis study suggests that instead of focusing solely on the finest image resolutions, multi-scale features should be emphasized for information extraction from high-resolution CXRs.
Schlagwörter
Stichwörter der Autor:innen
Autor:innen und Institutionen
- University of Tennessee, The Bredesen Center, Knoxville, Tennessee, United States.
- Oak Ridge National Laboratory, Computational Sciences and Engineering Division, Oak Ridge, Tennessee, United States.
- VA Connecticut Healthcare, West Haven, Connecticut, United States.
- VA Connecticut Healthcare System, Pain Research, Informatics, Multimorbidities, Education (PRIME) Center, West Haven, Connecticut, United States.
- Yale School of Medicine, Department of Medicine, New Haven, Connecticut, United States.
- Yale University, School of Public Health, New Haven, Connecticut, United States.
ORCID-Profile: Md Inzamam Ul Haque, Amy C Justice, Jacob D Hinkle
Förderung
- NCATS NIH HHS
Zitieren
APA
Haque, M. I. U., Dubey, A. K., Danciu, I., Justice, A. C., Ovchinnikova, O. S., & Hinkle, J. D. (2023). Effect of image resolution on automated classification of chest X-rays. Journal of Medical Imaging, 10(04), 1-16. https://doi.org/10.1117/1.jmi.10.4.044503
Vancouver
Haque MIU, Dubey AK, Danciu I, Justice AC, Ovchinnikova OS, Hinkle JD. Effect of image resolution on automated classification of chest X-rays. Journal of Medical Imaging. 2023;10(04):1-16. doi:10.1117/1.jmi.10.4.044503
BibTeX
@article{Haque2023,
author = {Haque, Md Inzamam Ul and Dubey, Abhishek K and Danciu, Ioana and Justice, Amy C and Ovchinnikova, Olga S and Hinkle, Jacob D},
title = {Effect of image resolution on automated classification of chest X-rays},
journal = {Journal of Medical Imaging},
year = {2023},
volume = {10},
number = {04},
pages = {1--16},
doi = {10.1117/1.jmi.10.4.044503}
}
EUVIMED