Skip to main content
OSTI.GOVU.S. Department of Energy Office of Scientific and Technical Information
U.S. Department of Energy
Office of Scientific and Technical Information
 

Effect of image resolution on automated classification of chest X-rays

Journal Article · · Journal of Medical Imaging
 [1];  [2];  [2];  [3];  [4];  [2]
  1. Univ. of Tennessee, Knoxville, TN (United States). Bredesen Center for Interdisciplinary Research and Graduate Education
  2. Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
  3. VA Connecticut Healthcare, West Haven, CT (United States); Yale Univ., New Haven, CT (United States)
  4. Univ. of Tennessee, Knoxville, TN (United States). Bredesen Center for Interdisciplinary Research and Graduate Education; Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
+ Show Author Affiliations
Deep 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. We 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. We 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. Lastly, 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.
Research Organization:
Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)
Sponsoring Organization:
USDOE Office of Science (SC)
Grant/Contract Number:
AC05-00OR22725
OSTI ID:
1996738
Journal Information:
Journal of Medical Imaging, Journal Name: Journal of Medical Imaging Journal Issue: 4 Vol. 10; ISSN 2329-4302
Publisher:
SPIECopyright Statement
Country of Publication:
United States
Language:
English

References (25)

Attention-Guided Curriculum Learning for Weakly Supervised Classification and Localization of Thoracic Diseases on Chest Radiographs book January 2018
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization journal October 2019
A review on lung boundary detection in chest X-rays journal February 2019
Exploring Large-scale Public Medical Image Datasets journal January 2020
Machine learning “red dot”: open-source, cloud, deep convolutional neural networks in chest radiograph binary normality classification journal September 2018
Deep learning journal May 2015
MIMIC-CXR, a de-identified publicly available database of chest radiographs with free-text reports journal December 2019
Detecting and classifying lesions in mammograms with Deep Learning journal March 2018
Deep Learning to Improve Breast Cancer Detection on Screening Mammography journal August 2019
Comparing different deep learning architectures for classification of chest radiographs journal August 2020
Automated abnormality classification of chest radiographs using deep convolutional neural networks journal May 2020
ImageNet: A large-scale hierarchical image database
  • Deng, Jia; Dong, Wei; Socher, Richard
  • 2009 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops (CVPR Workshops), 2009 IEEE Conference on Computer Vision and Pattern Recognition https://doi.org/10.1109/CVPR.2009.5206848
conference June 2009
Densely Connected Convolutional Networks conference July 2017
ChestX-Ray8: Hospital-Scale Chest X-Ray Database and Benchmarks on Weakly-Supervised Classification and Localization of Common Thorax Diseases conference July 2017
Thoracic Disease Identification and Localization with Limited Supervision conference June 2018
TieNet: Text-Image Embedding Network for Common Thorax Disease Classification and Reporting in Chest X-Rays conference June 2018
Hidden stratification causes clinically meaningful failures in machine learning for medical imaging conference April 2020
The U.S. Radiologist Workforce: An Analysis of Temporal and Geographic Variation by Using Large National Datasets journal April 2016
Deep Learning at Chest Radiography: Automated Classification of Pulmonary Tuberculosis by Using Convolutional Neural Networks journal August 2017
Development and Validation of Deep Learning–based Automatic Detection Algorithm for Malignant Pulmonary Nodules on Chest Radiographs journal January 2019
Assessment of Convolutional Neural Networks for Automated Classification of Chest Radiographs journal February 2019
The Effect of Image Resolution on Deep Learning in Radiography journal January 2020
Deep learning for chest radiograph diagnosis: A retrospective comparison of the CheXNeXt algorithm to practicing radiologists journal November 2018
CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison journal July 2019
Prediction of Obstructive Lung Disease from Chest Radiographs via Deep Learning Trained on Pulmonary Function Data journal January 2021

Similar Records

Domain Shift Analysis in Chest Radiographs Classification in a Veterans Healthcare Administration Population
Journal Article · Thu Apr 10 20:00:00 EDT 2025 · Journal of Imaging Informatics in Medicine · OSTI ID:2573574

Related Subjects

47 OTHER INSTRUMENTATION
60 APPLIED LIFE SCIENCES
chest X-ray
deep learning
image resolution
multitask classification
receptive field