Chest Radiograph Interpretation with Deep Learning Models: Assessment with Radiologist-adjudicated Reference Standards and Population-adjusted Evaluation
The DOI matches its Crossref record; 30 of the 30 references it cites were found in a registry or an index.
Abstract
Background Deep learning has the potential to augment the use of chest radiography in clinical radiology, but challenges include poor generalizability, spectrum bias, and difficulty comparing across studies. Purpose To develop and evaluate deep learning models for chest radiograph interpretation by using radiologist-adjudicated reference standards. Materials and Methods Deep learning models were developed to detect four findings (pneumothorax, opacity, nodule or mass, and fracture) on frontal chest radiographs. This retrospective study used two data sets. Data set 1 (DS1) consisted of 759 611 images from a multicity hospital network and ChestX-ray14 is a publicly available data set with 112 120 images. Natural language processing and expert review of a subset of images provided labels for 657 954 training images. Test sets consisted of 1818 and 1962 images from DS1 and ChestX-ray14, respectively. Reference standards were defined by radiologist-adjudicated image review. Performance was evaluated by area under the receiver operating characteristic curve analysis, sensitivity, specificity, and positive predictive value. Four radiologists reviewed test set images for performance comparison. Inverse probability weighting was applied to DS1 to account for positive radiograph enrichment and estimate population-level performance. Results In DS1, population-adjusted areas under the receiver operating characteristic curve for pneumothorax, nodule or mass, airspace opacity, and fracture were, respectively, 0.95 (95% confidence interval [CI]: 0.91, 0.99), 0.72 (95% CI: 0.66, 0.77), 0.91 (95% CI: 0.88, 0.93), and 0.86 (95% CI: 0.79, 0.92). With ChestX-ray14, areas under the receiver operating characteristic curve were 0.94 (95% CI: 0.93, 0.96), 0.91 (95% CI: 0.89, 0.93), 0.94 (95% CI: 0.93, 0.95), and 0.81 (95% CI: 0.75, 0.86), respectively. Conclusion Expert-level models for detecting clinically relevant chest radiograph findings were developed for this study by using adjudicated reference standards and with population-level performance estimation. Radiologist-adjudicated labels for 2412 ChestX-ray14 validation set images and 1962 test set images are provided. © RSNA, 2019 Online supplemental material is available for this article. See also the editorial by Chang in this issue.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“Only eight studies in ophthalmology imaging14,21,32,33,43,55,108,109, ten studies in respiratory imaging64,66,70,72,75,79,82,87,89,110 and six studies in breast imaging62,91,97,104,106,111 mentioned adherence to the STARD-2015 guidelines or had a STARD flow diagram in the manuscript.”
From Diagnostic accuracy of deep learning in medical imaging: a systematic review and meta-analysis (2021) · cites this paper as [82]
Citations
Cites
Fleischner Society: Glossary of Terms for Thoracic Imaging
Hansell, D. M., Bankier, A. A., MacMahon, H., McLoud, T. C., Müller, N. L., & Remy, J. (2008) · Radiology · doi:10.1148/radiol.2462070712 · Cited by 4,198 · cited 1×
“Clinical definitions for these categories were on the basis of the Fleischner Society Glos-sary of Terms for Thoracic Imaging, except for osseous fracture, which was defined as visible rib, clavicle, humeral, or vertebral body fractures (20).”
MethodArmato, S. G., McLennan, G., Bidaut, L., McNitt‐Gray, M. F., Meyer, C. R., Reeves, A. P., Zhao, B., Aberle, D. R., Henschke, C. I., Hoffman, E. A., et al. (2011) · Medical Physics · doi:10.1118/1.3528204 · Cited by 2,887
Revisiting Unreasonable Effectiveness of Data in Deep Learning Era
Sun, C., Shrivastava, A., Singh, S., & Gupta, A. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.97 · Cited by 2,484 · cited 1×
“We used Xception (22) as the convolutional neural network architecture that was pretrained on 300 000 000 natural images (23).”
MethodRajpurkar, P., Irvin, J., Ball, R. L., Zhu, K., Yang, B., Mehta, H., Duan, T., Ding, D., Bagul, A., Langlotz, C. P., et al. (2018) · PLOS Medicine · doi:10.1371/journal.pmed.1002686 · Cited by 1,449 · cited 2×
“Recent advances in deep learning offer promise for improving chest radiograph interpretation (1–4) and there are several recent reports of machine learning models achieving radiologist-level performance for different chest radiograph findings (5–7).”
BackgroundVery Deep Convolutional Networks for Text Classification
Conneau, A., Schwenk, H., Barrault, L., & Lecun, Y. (2017) · Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers · doi:10.18653/v1/e17-1104 · Cited by 1,046
Park, S. H., & Han, K. (2018) · Radiology · doi:10.1148/radiol.2017171920 · Cited by 894 · cited 1×
“However, because such data do not necessarily reflect real-world prevalence or diversity (33,34), such enrichment can also prevent meaningful clinical interpretation of diagnostic performance.”
BackgroundKrause, J., Gulshan, V., Rahimy, E., Karth, P., Widner, K., Corrado, G. S., Peng, L., & Webster, D. R. (2018) · Ophthalmology · doi:10.1016/j.ophtha.2018.01.034 · Cited by 564 · cited 1×
“…adjudicated labels, extensive work has been performed on the value of multiple interpretations in mammography (30) and other clinical settings (31,32). the most obviously positive and negative for pneumothorax, opacity, nodule or mass, or fracture for training, which can fail to address the…”
BackgroundSelf-citationMazurowski, M. A., Buda, M., Saha, A., & Bashir, M. R. (2019) · Journal of Magnetic Resonance Imaging · doi:10.1002/jmri.26534 · Cited by 554 · cited 1×
“Recent advances in deep learning offer promise for improving chest radiograph interpretation (1–4) and there are several recent reports of machine learning models achieving radiologist-level performance for different chest radiograph findings (5–7).”
BackgroundNam, J. G., Park, S., Hwang, E. J., Lee, J. H., Jin, K.-N., Lim, K. Y., Vu, T. H., Sohn, J. H., Hwang, S., Goo, J. M., et al. (2019) · Radiology · doi:10.1148/radiol.2018180237 · Cited by 539 · cited 1×
“Recent advances in deep learning offer promise for improving chest radiograph interpretation (1–4) and there are several recent reports of machine learning models achieving radiologist-level performance for different chest radiograph findings (5–7).”
BackgroundMcBee, M. P., Awan, O. A., Colucci, A. T., Ghobadi, C. W., Kadom, N., Kansagra, A. P., Tridandapani, S., & Auffermann, W. F. (2018) · Academic Radiology · doi:10.1016/j.acra.2018.02.018 · Cited by 496
Research works printed but not linked: 3
- Auffermann WF, 2019, AJR Am J Roentgenol, 1
- Summers RM, NIH Chest X-ray Dataset of 14 Common Thorax Disease Categories
- Chollet F. Xception: Deep Learning with Depthwise Separable Convolutions. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Honolulu, July 21–26, 2017. Piscataway, NJ: IEEE, 2017; 1800–1807.
Cited by
Items 301–337 of 337
Andersson, P., Johnsson, J., Björnsson, O., Cronberg, T., Hassager, C., Zetterberg, H., Stammet, P., Undén, J., Kjaergaard, J., Friberg, H., et al. (2021) · Figshare · doi:10.6084/m9.figshare.14121656
Andersson, P., Johnsson, J., Björnsson, O., Cronberg, T., Hassager, C., Zetterberg, H., Stammet, P., Undén, J., Kjaergaard, J., Friberg, H., et al. (2021) · Figshare · doi:10.6084/m9.figshare.14121650
Using Machine Learning Towards Decision Support for Refractory Epilepsy Cases
Farhoudi, B. (2023) · Libraries and Cultural Resources (University of Calgary) · doi:10.11575/prism/40701
Andersson, P., Johnsson, J., Björnsson, O., Cronberg, T., Hassager, C., Zetterberg, H., Stammet, P., Undén, J., Kjaergaard, J., Friberg, H., et al. (2021) · Figshare · doi:10.6084/m9.figshare.14121662
Ligduurvoorspelling voor Ziekenhuismanagement
Wamba Momo, L. N. (2024) · Lirias (KU Leuven)
Penerapan Convolutional Neural Network Untuk Klasifikasi Citra Penyakit Kardiomegali
Rizalde, A. R. (2025) · Repository UIN Sultan Syarif Kasim Riau (Universitas Islam Negeri Sultan Syarif Kasim Riau)
Artificial Intelligence-based Decision Support Models for COVID-19 Detection
Pereira, S. C. (2024) · Portuguese National Funding Agency for Science, Research and Technology (RCAAP Project by FCT) · doi:10.34626/dqz3-2b46
Classification of Chest XRay Diseases through image processing and analysis techniquesPreprint
Novoa, S. M., Ibáñez, M. C., Mesa, L. G., & Kramer, J. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2601.10913
“[6] The inherent ambiguity of CXR presentations poses a diagnostic hurdle even for experienced healthcare professionals.”
Bhagi, C., Gudavalli, M. R., Kruse, R. A., & Cox, J. M. (2026) · Bioengineering · doi:10.3390/bioengineering13030330
“Recent advances in AI-based analysis of medical images have primarily focused on single imaging modalities, such as X-ray or ultrasound [27,28].”
Optimizing Chest Disease Diagnosis: Machine Learning Classification of Critical Parameters
Aggarwal, R., Gupta, R., & Jain, C. (2025) · International Conference on Advanced Computing and Emerging Technologies (ACET) · doi:10.1109/acet67282.2025.11430289
“The study known as Chest Radiograph Interpretation [3] was centered on deep learning, which has the ability to match or even exceed the efficiency of Radiologists in the diagnosis of the chest X-ray.”
Kim, M. Y., Yoo, J. Y., Jang, Y. H., Hong, J. H., Lee, S. H., Choi, Y. R., & Jin, K. N. (2026) · Journal of the Korean Society of Radiology · doi:10.3348/jksr.2025.0071
“These are referred to as actionable abnormalities (AAs), which account for approximately 1.5%–32.5% of radiology reports (2, 5, 6).”
InfluentialRabby, A., Jannat, J. N., Assaduzzaman, M., Sarker, R. K. R., & Tusher, R. T. H. (2026) · Bulletin of Electrical Engineering and Informatics · doi:10.11591/eei.v15i2.10084
Zeenat Akram, U.-E.-A. S., Sabir, U.-E.-A., Atif, A., Khalil, A., Akhtar, M. A., & Rafiq, M. (2026) · Pakistan Journal of Medical & Cardiological Review · doi:10.5281/zenodo.20049004
Kim, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2605.19201
“Variants such as fine-tuned VGG16 [15] outperformed state-of-the-art deep learning algorithms such as ResNet-50 in pneumonia classification task, and Xception-based models [16] extended this approach to broader thoracic findings, achieving performance comparable to board-certified radiologists.”
Lung Disease Classification through Explainable AI and Transfer Learning Approach
Srivastava, A. K., Verma, G., Ikra, K., Ahmad, S. K., Shivahare, B. D., & Prakash, S. (2026) · International Conference on Intelligent Processing, Hardware, Electronics, and Radio Systems (CIPHER) · doi:10.1109/cipher70417.2026.11523889
Kim, D. (2026) · IEEE Access · doi:10.1109/access.2026.3695715
“Follow-up models, including fine-tuned VGG16 [18] and Xception-based architectures [19], further improved pneumonia and thoracic-disease classification, achieving performance comparable to that of board-certified radiologists.”
Kohzai, M., Yamamoto, S., Matsushita, M., Ueno, Y., & Tanigawa, N. (2026) · Diagnostics · doi:10.3390/diagnostics16111637
Analysis of Weld CT Images Using Neural Networks
Jurkovič, M. (2026) · Brno University of Technology Digital Library (Brno University of Technology)
Thota, S. K., Alenezi, F. S., Polat, K., & Alhudhaif, A. (2026) · Applied Soft Computing · doi:10.1016/j.asoc.2026.115787
Paracha, M. F., Mahmood, M., & Farhan, M. (2026) · Radio Electronics Computer Science Control · doi:10.15588/1607-3274-2026-2-9
Cethraian-X: A Leakage-Clean, Multi-Seed Benchmark of Chest X-Ray Classification Under Weak LabelsPreprint
Badhan, M. (2026) · Research Square · doi:10.21203/rs.3.rs-10484985/v1
Figueredo, M. F. B., Alves, N. D. P. D. C., Miranda, W. L. S., & Simeone, D. (2026) · Discover Artificial Intelligence · doi:10.1007/s44163-026-01910-z
“Traditionally, the interpretation of imaging examinations, such as radiography, computed tomography, magnetic resonance imaging, and mammography, relies on professional expertise, requiring not only technical knowledge but also extensive clinical experience [4].”
Positive-Unlabeled Preference Optimization For Chest X-ray Report GenerationPreprint
Kobayashi, Y., Ramesh, P., Chaudhry, M., Jeanselme, V., Gichoya, J. W., Koyejo, S., Capaccione, K. M., & Joshi, S. D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.05341
“…report generation as one-sided noise in which positive visual findings are unlabeled in retrospective reports, not because of diagnostic uncertainty or interpretative disagreement but due to clinical workflow and prioritization [Rudolph et al., 2021, Majkowska et al., 2020, Jain et al., 2021b].”
Taşkıran, D., Aslan, S., Kolsuz, S., Alçı, M., & Yiğitbaş, E. Y. (2026) · Diagnostics · doi:10.3390/diagnostics16162576
Selective multipathology chest X-ray classification via rejection mechanisms
Aperstein, Y., Tzahar, A., Gottlib, A., Verber, T., Damti, R. S., & Apartsin, A. (2026) · Scientific Reports · doi:10.1038/s41598-026-66294-7
Wang, Y., Yang, J., & Zhang, H. (2026) · Fractal and Fractional · doi:10.3390/fractalfract10090597
Zhao, L., Chen, L., Zhao, F., Li, J., Zhang, H., Zhou, Y., Li, J., Jin, R., Zhang, S., Liu, Y., et al. (2026) · Journal of Thoracic Disease · doi:10.21037/jtd-2026-0914
“However, these methods have notable limitations: substantial inter-observer variability, need for specialized experience in image interpretation, and insufficient sensitivity to early lesions, when intervention may be most effective (8).”
Self-citationIşık, Ş., & Eren, H. A. (2026) · Tomography · doi:10.3390/tomography12090129
“Large labeled resources such as CheXpert and MIMIC-CXR/MIMIC-CXR-JPG, distributed through the PhysioNet platform, have enabled models for image-level and finding-level classification [3–6], and several systems have reported performance comparable to radiologists for selected findings [7,8].”
Duais, F. S. A., Elebeed, R. A. E., El-Aziz, R. M. A., & Taloba, A. I. (2026) · Natural and Engineering Sciences · doi:10.28978/nesciences.263045
Large Language Models Meet Gynecologic Ultrasound: Advancing the Characterization of ADNEXal Masses
Soccio, G., Napoli, S. D., Trerotoli, P., Loizzi, V., Zompì, L. G., Colonna, G., Forgia, D. L., Cormio, G., & Arezzo, F. (2026) · Journal of Imaging · doi:10.3390/jimaging12090462
“Multiple studies have demonstrated its validity in the fields of screening [9,10], triage [11,12], diagnosis [13,14], drug development [15,16], treatment [17,18], monitoring [19], and image interpretation [20,21].”
Damar, Ö., & Yaman, M. (2026) · Tomography · doi:10.3390/tomography12100142
“Target-level heterogeneity in recent emergency and commercial-AI evaluations further supports this interpretation [19,20,26].”
Machine learning algorithm-based decision support system for prime bank stock trend prediction
Gunawan, G., Budiono, W., Andriani, W., & Naja, N. N. P. W. (2024) · Journal of Intelligent Decision Support System (IDSS)
Patel, K., Beedala, P., Vora, D., & Mehta, H. (2026) · medRxiv · doi:10.64898/2026.10.03.26364646
When Does Data Augmentation Help with Few Chest-Radiograph Training Patients?
Sariyar, M. (2026) · Mathematics · doi:10.3390/math14203732
Dahiru, I., Adam, B. A., Umar, S. N., & Pal, B. L. (2026) · Research Square · doi:10.21203/rs.3.rs-11444323/v1
Integration of Artificial Intelligence, Healthcare, and Information Technology
Indhumathi, G., Saranya, G., Fletcher, A. S. A., Kumar, R. N., Ajayan, J., & Sreejith, S. (2026) · Wiley eBooks · doi:10.1002/9781394302574.ch2
Supporting information 10 of AI-based radiodiagnosis using chest X-rays: A review
Akhter, Y., Singh, R., & Vatsa, M. (2023) · doi:10.3389/fdata.2023.1120989.s010
Metadata
- Authors
- Majkowska, A., Mittal, S., Steiner, D. F., Reicher, J. J., McKinney, S. M., Duggan, G. E., Eswaran, K., Cameron Chen, P.-H., Liu, Y., Kalidindi, S. R., Ding, A., Corrado, G. S., Tse, D., & Shetty, S.
- Affiliations
- Stanford Health Care (US)
- From Google Health, 1600 Amphitheatre Pkwy, Mountain View, CA 94043 (A.M., S.M., D.F.S., S.M.M., G.E.D., K.E., P.H.C.C., Y.L., G.S.C., D.T., S.S.); Stanford Healthcare and Palo Alto Veterans Affairs, Palo Alto, Calif (J.J.R.); Apollo Radiology International, Hyderabad, India (S.R.K.); and California Advanced Imaging, Novato, Calif (A.D.).
- ORCID
- Venue
- Radiology
- Published
- 2020-02
- Volume
- 294
- Issue
- 2
- Pages
- 421-431
- ISSN
- 0033-8419 (ISSN-L)
- 1527-1315
- OpenAlex
- W2991676991
- Cited by
- 277
- References
- 30
- Citations by year
- 2026: 21 · 2025: 37 · 2024: 36 · 2023: 48 · 2022: 48 · 2021: 52 · 2020: 34
- FWCI
- 22.28
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Radiology (publishedVersion)
- PubMed (publishedVersion)
- Topics
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › COVID-19 diagnosis using AI
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Health Sciences › Medicine › Radiology, Nuclear Medicine and Imaging › Radiomics and Machine Learning in Medical Imaging
- Keywords
- chest X-ray interpretation, deep learning, pneumothorax detection, lung nodule detection, lung opacity, rib fracture detection, inverse probability weighting, spectrum bias, generalizability, ChestX-ray14, positive predictive value, sensitivity and specificity
- MeSH terms
- Radiologists, Deep Learning, Adolescent, Adult, Aged, Aged, 80 and over, Child, Child, Preschool, Female, Humans, Infant, Male, Middle Aged, Pneumothorax, Radiographic Image Interpretation, Computer-Assisted, Reference Standards, Respiratory Tract Diseases, Retrospective Studies, Sensitivity and Specificity, Thoracic Injuries, Radiography, Thoracic, Reproducibility of Results, Young Adult
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.