Overview

Schema Version

https://atlas.rsna.org/schemas/2025-11/dataset.json

Name

Performance and Usability of Code-Free Deep Learning for Chest Radiograph Classification, Object Detection, and Segmentation

Link

https://doi.org/10.1148/ryai.220062

Indexing

Keywords: Chest radiograph, Code-free deep learning, Automated machine learning, Pneumonia, Pneumothorax, Multilabel classification, Object detection, Segmentation, NIH-CXR14, CheXpert, RSNA Pneumonia, SIIM-ACR

Content: CH, IN, RS

RadLex: RID28518, RID5335, RID5352

SNOMED: 36118008, 8186001, 46621007, 233604007

Author(s)

Samantha M. Santomartino

Nima Hafezi-Nejad

Vishwa S. Parekh

Paul H. Yi

Organization(s)

University of Maryland Medical Intelligent Imaging (UM2ii) Center, Department of Diagnostic Radiology and Nuclear Medicine, University of Maryland School of Medicine

The Russell H. Morgan Department of Radiology and Radiological Science, Johns Hopkins University School of Medicine

Department of Computer Science, Whiting School of Engineering, Johns Hopkins University

Malone Center for Engineering in Healthcare, Johns Hopkins University

License

URL: https://pubs.rsna.org/doi/10.1148/ryai.220062

Contact

Corresponding author: Paul H. Yi, email: ude.dnalyramu.mos@iyp

Funding

Amazon Web Services granted a proof-of-concept credit for investigation of the Amazon platform only.

Ethical review

All images were de-identified, from public open-access databases; no institutional review board approval was required; HIPAA compliant.

Comments

Study evaluating six code-free deep learning platforms on public chest radiograph datasets for classification, object detection, and segmentation.

Date

Published: 2023-02-15

References

[1] Santomartino SM, Hafezi-Nejad N, Parekh VS, Yi PH. "Performance and Usability of Code-Free Deep Learning for Chest Radiograph Classification, Object Detection, and Segmentation". Radiology: Artificial Intelligence. 2023-03-01. doi:10.1148/ryai.220062. PMID: 37035428. PMCID: PMC10077092.

[2] Kermany D. "Large dataset of labeled optical coherence tomography (OCT) and Chest X-Ray images". Mendeley Data. .

[3] Wang X, Peng Y, Lu L, Lu Z, Bagheri M, Summers RM. "ChestX-ray8: Hospital-scale chest x-ray database and benchmarks on weakly-supervised classification and localization of common thorax diseases". CVPR 2017. . Available from: http://openaccess.thecvf.com/content_cvpr_2017/html/Wang_ChestX-ray8_Hospital-Scale_Chest_CVPR_2017_paper.html

[4] Irvin J, Rajpurkar P, Ko M, et al.. "CheXpert: A Large Chest Radiograph Dataset with Uncertainty Labels and Expert Comparison". AAAI 2019. .

[5] . "RSNA Pneumonia Detection Challenge". Kaggle. . Available from: https://www.kaggle.com/c/rsna-pneumonia-detection-challenge

[6] . "SIIM-ACR Pneumothorax Segmentation". Kaggle. . Available from: https://www.kaggle.com/c/siim-acr-pneumothorax-segmentation/overview/description

Dataset

Motivation

To evaluate performance and usability of code-free deep learning platforms for chest radiograph analysis across classification, object detection, and segmentation tasks.

Sampling

Retrospective selection of publicly available chest radiograph datasets widely used in literature for the targeted tasks.

Partitioning scheme

Random split of data into 80% training, 10% validation, and 10% testing when manual designation was allowable; preserved publisher-provided splits when applicable.

Missing information

Demographics were not reported for Guangzhou, RSNA, or SIIM datasets in this study; image file formats and preprocessing steps not reported.

External data

External testing performed using subsets of NIH-CXR14 (pediatric subset, 383 images) and CheXpert (attempted 151,522 images; crashed after 4763).

Confidentiality

De-identified images from public, open-access datasets; HIPAA compliant.

Re-identification

Not applicable; source datasets were de-identified.

Sensitive data

No PHI; datasets are de-identified public collections.